Driving mode switching method and vehicle

Abstract

This application provides a driving mode switching method and a vehicle. The method includes: acquiring multi-source parameters of the vehicle, wherein the parameter types of the multi-source parameters include navigation parameters and state perception parameters; performing weighted fusion analysis on the multi-source parameters based on multiple evaluation dimensions to obtain fusion analysis results, wherein the fusion analysis results are used to characterize the fusion features of the vehicle in the current driving environment; determining a target driving mode from multiple candidate driving modes based on the fusion analysis results; performing navigation prediction analysis using the multi-source parameters to determine the switching timing corresponding to the target driving mode; and sending mode switching commands to multiple control units of the vehicle according to the target driving mode and the switching timing to control the vehicle to switch to the target driving mode. This application solves the technical problem of poor system safety and stability caused by single parameters and unreasonable switching timing in existing driving mode switching schemes.

Description

Technical Field

[0001] This application relates to the field of vehicles, and more specifically, to a driving mode switching method and a vehicle. Background Technology

[0002] Automated passenger vehicle driving mode control technology is mainly divided into two categories: manual switching and preliminary automatic switching. Manual switching is highly susceptible to causing vehicle loss of control or inaccurate maneuvering, posing a high safety risk, and requires a certain level of user skill. Preliminary automatic switching suffers from problems such as limited decision parameters, low switching accuracy, lack of predictive capability, passive response to changing scenarios, imperfect parameter fusion logic, and weak anti-interference ability. Therefore, existing vehicle driving mode switching solutions exhibit poor system safety and stability. Currently, there is no satisfactory solution to these problems. Summary of the Invention

[0003] This application provides a driving mode switching method and vehicle to at least solve the technical problems of poor system safety and stability caused by single parameters and unreasonable switching timing in existing driving mode switching schemes.

[0004] According to one aspect of the embodiments of this application, a driving mode switching method is provided. The method includes: acquiring multi-source parameters of a vehicle, wherein the parameter types of the multi-source parameters include navigation parameters and state perception parameters; performing weighted fusion analysis on the multi-source parameters based on multiple evaluation dimensions to obtain a fusion analysis result, wherein the fusion analysis result is used to characterize the fusion features of the vehicle in the current driving environment; determining a target driving mode from multiple candidate driving modes based on the fusion analysis result; performing navigation prediction analysis using the multi-source parameters to determine the switching timing corresponding to the target driving mode; and sending mode switching commands to multiple control units of the vehicle according to the target driving mode and the switching timing to control the vehicle to switch to the target driving mode.

[0005] Furthermore, a weighted fusion analysis is performed on the multi-source parameters based on multiple evaluation dimensions to obtain the fusion analysis results, including: preprocessing the multi-source parameters to obtain a preprocessed parameter set; determining the current driving scenario type of the vehicle based on the preprocessed parameter set; determining the scenario correction coefficients corresponding to each of the multiple evaluation dimensions according to the driving scenario type; calculating the dimension weights corresponding to each of the multiple evaluation dimensions using the scenario correction coefficients and the basic weights corresponding to each of the multiple evaluation dimensions; and weighting and accumulating the preprocessed parameter set using the dimension weights to obtain the fusion analysis results.

[0006] Furthermore, the multi-source parameters are preprocessed to obtain a preprocessed parameter set, including: validating the multi-source parameters to obtain a validation result; when the validation result shows that all multi-source parameters are valid, normalizing the multi-source parameters to obtain a preprocessed parameter set; when the validation result shows that the navigation parameters are invalid, using the real-time terrain parameters from the state-aware parameters to replace the navigation parameters to obtain a preprocessed parameter set; when the validation result shows that the road surface adhesion coefficient from the state-aware parameters is invalid, using at least one of the environmental warning information, road surface friction state information, and environmental temperature information from the state-aware parameters to replace the road surface adhesion coefficient to obtain a preprocessed parameter set; when the validation result shows that a single sensor parameter from the state-aware parameters is invalid, using other environmental parameters from the state-aware parameters besides the single sensor parameter for cross-validation to obtain a validation result, and using the validation result to replace the single sensor parameter to obtain a preprocessed parameter set.

[0007] Furthermore, based on the preprocessed parameter set, determining the driving scenario type includes: extracting driving scenario determination parameters from the preprocessed parameter set, wherein the driving scenario determination parameters include at least one of the following: environmental warning information, road surface adhesion coefficient, long slope identification, slope information, road type information, and vehicle speed information; matching the driving scenario determination parameters with preset scenario determination rules to determine the driving scenario type, wherein the driving scenario type includes at least one of the following: icy and slippery scenario, long downhill and steep slope scenario, highway straight road scenario, urban congestion scenario, and ordinary scenario.

[0008] Furthermore, the multiple assessment dimensions include safety assessment, terrain assessment, energy consumption assessment, and user habit assessment. Based on the driving scenario type, the scenario correction coefficients for each assessment dimension are determined as follows: When the driving scenario is an icy or slippery scenario, the scenario correction coefficient for the safety assessment dimension is determined as the first value, and the scenario correction coefficients for the terrain assessment, energy consumption assessment, and user habit assessment dimensions are determined as the second value, where the first value is greater than the second value; when the driving scenario is a long downhill or steep slope scenario, the scenario correction coefficient for the terrain assessment dimension is determined as the third value, and the scenario correction coefficients for the safety assessment, energy consumption assessment, and user habit assessment dimensions are determined as... The fourth value is determined by the fact that the third value is greater than the fourth value. When the driving scenario is a highway straight road scenario, the scenario correction coefficients for the energy consumption assessment dimension and the user habit assessment dimension are determined as the fifth value, and the scenario correction coefficients for the safety assessment dimension and the terrain assessment dimension are determined as the sixth value, where the fifth value is greater than the sixth value. When the driving scenario is an urban congestion scenario, the scenario correction coefficient for the user habit assessment dimension is determined as the seventh value, and the scenario correction coefficients for the safety assessment dimension, the terrain assessment dimension, and the energy consumption assessment dimension are determined as the eighth value, where the seventh value is greater than the eighth value. When the driving scenario is a normal scenario, the scenario correction coefficients corresponding to multiple assessment dimensions are determined as the ninth value.

[0009] Furthermore, based on the fusion analysis results, determining the target driving mode from multiple candidate driving modes includes: mapping the fusion analysis results to fuzzy linguistic variables according to the numerical range corresponding to the fusion analysis results; using preset pattern reasoning rules to reason about the fuzzy linguistic variables to determine the membership degrees corresponding to the multiple candidate driving modes; and selecting the target driving mode from the multiple candidate driving modes according to the membership degrees.

[0010] Furthermore, the fusion analysis results include fusion values ​​for safety assessment dimension, terrain assessment dimension, and energy consumption assessment dimension; multiple candidate driving modes include snow mode, comfort mode, economy mode, and sport mode; using pattern inference rules to infer fuzzy linguistic variables, the membership degrees corresponding to the multiple candidate driving modes are determined, including mapping the safety assessment dimension fusion values ​​to safety level linguistic variables, mapping the terrain assessment dimension fusion values ​​to terrain level linguistic variables, and mapping the energy consumption assessment dimension fusion values ​​to energy consumption level linguistic variables; and using pattern inference rules to infer the safety level linguistic variables, terrain level linguistic variables, energy consumption level linguistic variables, and user habit parameters to determine the membership degrees corresponding to snow mode, comfort mode, economy mode, and sport mode.

[0011] Furthermore, navigation prediction analysis using multi-source parameters is used to determine the switching timing, including: extracting the predicted distance corresponding to the target road segment from navigation parameters; extracting the vehicle's current real-time speed from state perception parameters; and determining the switching timing based on the predicted distance, real-time speed, and preset mode smooth transition duration.

[0012] Furthermore, based on the target driving mode and the switching timing, sending mode switching commands to multiple control units includes: determining parameter adjustment information corresponding to multiple control units based on the target driving mode and the switching timing, wherein the multiple control units include: powertrain control unit, body control system, suspension control system and steering system; and sending mode switching commands to multiple control units respectively based on the parameter adjustment information to control the multiple control units to perform parameter adjustment actions in a linear and gradual manner.

[0013] According to another aspect of the embodiments of this application, a control device for a driving mode switching system is also provided, comprising: an acquisition module for acquiring multi-source parameters of a vehicle, wherein the parameter types of the multi-source parameters include navigation parameters and state perception parameters; an analysis module for performing weighted fusion analysis on the multi-source parameters based on multiple evaluation dimensions to obtain a fusion analysis result, wherein the fusion analysis result is used to characterize the fusion features of the vehicle in the current driving environment; a first determination module for determining a target driving mode from multiple candidate driving modes based on the fusion analysis result; a second determination module for performing navigation prediction analysis using the multi-source parameters to determine the switching timing corresponding to the target driving mode; and a control module for sending mode switching commands to multiple control units of the vehicle according to the target driving mode and the switching timing to control the vehicle to switch to the target driving mode.

[0014] According to another aspect of the embodiments of this application, a vehicle is also provided, including: a memory storing an executable program; and a processor for running the program, wherein the program executes the methods in various embodiments of this application when it runs.

[0015] According to another aspect of the embodiments of this application, a computer-readable storage medium is also provided, the computer-readable storage medium including a stored executable program, wherein, when the executable program is running, it controls the device where the computer-readable storage medium is located to perform the methods of various embodiments of this application.

[0016] According to another aspect of the embodiments of this application, a computer program product is also provided, including a computer program that, when executed by a processor, implements the methods of various embodiments of this application.

[0017] According to another aspect of the embodiments of this application, a computer program product is also provided, including a non-volatile computer-readable storage medium storing a computer program, which, when executed by a processor, implements the methods in various embodiments of this application.

[0018] According to another aspect of the embodiments of this application, a computer program is also provided, which, when executed by a processor, implements the methods of the various embodiments of this application.

[0019] In this embodiment, a driving mode switching method is employed. This method acquires multi-source parameters of the vehicle, including navigation parameters and state perception parameters. It performs weighted fusion analysis on the multi-source parameters based on multiple evaluation dimensions to obtain fusion analysis results, which characterize the vehicle's fusion features in the current driving environment. Based on the fusion analysis results, a target driving mode is determined from multiple candidate driving modes. Navigation prediction analysis is performed using the multi-source parameters to determine the switching timing corresponding to the target driving mode. Based on the target driving mode and the switching timing, mode switching commands are sent to multiple control units of the vehicle to control the vehicle to switch to the target driving mode. This application solves the technical problem of poor system safety and stability caused by single parameters and unreasonable switching timing in existing driving mode switching schemes. It achieves the goals of pre-planning switching timing, accurately matching driving modes using dynamic weighted fusion of multiple parameters, and executing switching through smooth transition control. This results in improved accuracy and safety of mode switching, elimination of switching jerks, and optimized driving comfort. Attached Figure Description

[0020] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0021] Figure 1 This is a flowchart of a driving mode switching method according to an embodiment of this application;

[0022] Figure 2 This is a flowchart illustrating an automatic switching technology for passenger vehicle driving modes based on navigation prediction and multi-parameter fusion, according to an embodiment of this application.

[0023] Figure 3 This is a structural block diagram of a driving mode switching system according to an embodiment of this application. Detailed Implementation

[0024] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0025] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0026] According to an embodiment of this application, a driving mode switching method embodiment is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Also, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0027] This embodiment provides a driving mode switching method. Figure 1 This is a flowchart of a driving mode switching method according to an embodiment of this application, such as... Figure 1 As shown, the process includes the following steps:

[0028] Step S11: Obtain multi-source parameters of the vehicle, wherein the parameter types of the multi-source parameters include navigation parameters and state perception parameters.

[0029] In this embodiment of the application, the driving mode switching system acquires multi-source parameters of the vehicle. The parameter types of the multi-source parameters include navigation parameters and state perception parameters. The multi-source parameters are a comprehensive data set used to describe the current driving state of the vehicle and the road environment ahead. They integrate complementary information from different sources and of different natures, and are used to eliminate blind spots or errors of a single sensor through data complementarity.

[0030] Navigation parameters are static or semi-static road feature data with a forward-looking perspective, obtained from high-precision maps or real-time navigation services. They mainly reflect the predictive information on road geometry, pavement attributes, and environmental warnings within a certain distance ahead.

[0031] State perception parameters are dynamic vehicle operating parameters and real-time environmental parameters collected in real time by various on-board physical sensors. They mainly reflect the vehicle's current dynamic state, load status, and real-time surrounding micro-environment conditions. Sensors may include cameras, rain sensors, temperature sensors, and millimeter-wave radar, etc.

[0032] For example, when a vehicle is about to enter a long downhill section in a mountainous area, the system first obtains the prediction information from the navigation parameters that the slope (or average slope) within 1 kilometer ahead is negative, the long slope is marked as valid, and the curve curvature is large. At the same time, it reads the real-time data from the state perception parameters that the current vehicle speed is 80 km / h, the longitudinal acceleration is negative, the road surface adhesion coefficient is estimated to be 0.65, and the brake pedal pressure is 0.1 MPa. The system integrates these two sets of data from different time spans into a feature vector that includes terrain, dynamics, road surface adhesion, and other dimensions, as the input basis for the next step of weighted fusion analysis.

[0033] Therefore, by integrating the long-distance prediction capability of navigation with the real-time high-precision perception capability of sensors, a data foundation is provided for subsequent calculation of accurate pattern matching and smooth switching timing based on multi-parameter fusion.

[0034] Step S12: Perform weighted fusion analysis on multi-source parameters based on multiple evaluation dimensions to obtain fusion analysis results. The fusion analysis results are used to characterize the fusion features of the vehicle in the current driving environment.

[0035] In this embodiment of the application, the driving mode switching system performs weighted fusion analysis on multi-source parameters based on multiple evaluation dimensions to obtain fusion analysis results. The evaluation dimensions are logical levels used to classify and prioritize multi-source parameters.

[0036] The fusion analysis result refers to the comprehensive feature values ​​under various dimensions obtained after weighted calculation. This result quantifies the comprehensive form characteristics of the current vehicle and provides data support for subsequent pattern matching.

[0037] For example, multiple evaluation dimensions may include: security dimensions, terrain dimensions, energy consumption dimensions, and user habit dimensions.

[0038] Therefore, a dynamic weight allocation mechanism is introduced, which changes the rigidity of traditional fixed-weight decision-making. This allows the system to adaptively adjust the influence of each parameter according to the urgency and key features of different driving scenarios, significantly improving the accuracy of pattern matching.

[0039] Step S13: Based on the fusion analysis results, determine the target driving mode from multiple candidate driving modes.

[0040] In this embodiment, the driving mode switching system determines the target driving mode from multiple candidate driving modes based on the fusion analysis results. The multiple candidate driving modes are a set of switchable driving modes preset by the system, such as Eco mode, Comfort mode, Sport mode, and Snow mode, each corresponding to specific parameters such as power response, suspension stiffness, and energy recovery.

[0041] The target driving mode refers to the driving mode that is most suitable for the current driving environment, which is finally selected based on the results of the fusion analysis and by comparing the matching degree of each candidate mode.

[0042] For example, multiple candidate driving modes may include: Eco mode (emphasizing low energy consumption), Comfort mode (emphasizing smoothness), Sport mode (emphasizing power response) and Snow mode (emphasizing safety).

[0043] Therefore, by determining the target model through quantitative evaluation based on the results of fusion analysis, the blindness of subjective judgment or single-parameter decision-making is avoided, and the model selection is highly matched with the current actual working conditions.

[0044] Step S14: Use multi-source parameters to perform navigation prediction analysis and determine the switching timing corresponding to the target driving mode.

[0045] In this embodiment, the driving mode switching system uses multi-source parameters for navigation prediction analysis to determine the switching timing corresponding to the target driving mode.

[0046] Navigation prediction analysis refers to the process of using road geometry information and environmental warning information provided by in-vehicle navigation systems or high-precision maps, combined with real-time vehicle kinematic data, to perform forward-looking calculations and assessments of the characteristics of specific road sections ahead.

[0047] The timing of the mode switch refers to the specific time or distance point at which the system calculates the start time or distance for the mode switch command to ensure a smooth transition of the driving mode before the vehicle enters the target road segment. This timing does not refer to the moment when the mode switch is completed, but rather to the starting moment when the switching action begins, in order to allow sufficient time for parameter adjustment to eliminate the feeling of jerkiness.

[0048] For example, taking a snowy road section as an example, the navigation prediction analysis shows that there is a snowy road section 500 meters ahead, that is, the predicted distance is 500 meters. At this time, the vehicle's real-time speed is 60 km / h (equivalent to 16.7 m / s). The system sets the mode smooth transition time to 1 second. According to the formula calculation, the switching timing is 31 seconds. This means that the system determines to start the mode switching process at a distance of 500 meters from the snowy road section (i.e., 31 seconds from the current moment) to ensure that when the vehicle reaches the snowy entrance, the driving mode has been smoothly switched to snow mode.

[0049] Therefore, by introducing a switching timing prediction mechanism based on distance and speed calculations, the problem of mode mismatch after entering the target road segment caused by reaction lag in traditional driving mode switching is solved.

[0050] Step S15: Based on the target driving mode and the timing of the switch, send mode switching commands to multiple control units of the vehicle to control the vehicle to switch to the target driving mode.

[0051] In this embodiment, the driving mode switching system sends mode switching commands to multiple control units of the vehicle according to the target driving mode and the switching timing, so as to control the vehicle to switch to the target driving mode.

[0052] The mode switching command refers to a control signal that includes the target mode identifier and the specific parameter adjustment target, which is used to guide the vehicle's various subsystems to perform the corresponding parameter changes.

[0053] Multiple control units refer to electronic control devices in a vehicle that are responsible for implementing different functional modules, including powertrain control unit, body control unit, suspension control unit and steering system, which are respectively responsible for adjusting functions such as power output, vehicle stability, suspension stiffness and steering assist.

[0054] For example, when the system clock strikes or the vehicle position reaches the switching time, the system generates a corresponding mode switching command based on the target driving mode. This command is sent to relevant control units within the vehicle, such as the powertrain control unit and the body control unit. The command specifies the parameter adjustment targets for each control unit in the target mode; for example, the powertrain control unit adjusts the power response delay time, and the body control unit adjusts the suspension stiffness and steering assist. After receiving the command, each control unit gradually adjusts the current parameters to the set values ​​of the target mode according to a preset smooth transition strategy (such as linear gradual change) within a specified smooth transition time, thereby completing the switch from the current mode to the target driving mode.

[0055] Thus, by distributing instructions based on precise switching timing, early intervention and smooth execution of driving mode switching are achieved, effectively avoiding vehicle handling discomfort or jerking caused by delayed mode switching.

[0056] In summary, the driving mode switching method of this application first acquires multi-source parameters of the vehicle, including navigation parameters and state perception parameters. By integrating the long-distance prediction capability of navigation with the real-time high-precision perception capability of sensors, a data foundation is provided for subsequent accurate mode matching and smooth switching timing calculation based on multi-parameter fusion. Then, a weighted fusion analysis is performed on the multi-source parameters based on multiple evaluation dimensions to obtain the fusion analysis results. These results characterize the vehicle's fusion features in the current driving environment. This introduces a dynamic weight allocation mechanism, overcoming the rigidity of traditional fixed-weight decision-making. The system can adaptively adjust the influence of each parameter according to the urgency and feature emphasis of different driving scenarios, significantly improving the accuracy of mode matching. Secondly, based on the fusion analysis results, multiple candidate driving modes are selected... The target driving mode is determined through quantitative evaluation based on fusion analysis results, avoiding the blindness of subjective judgment or single-parameter decision-making and ensuring a high degree of matching between the mode selection and the current actual operating conditions. Next, navigation prediction analysis is performed using multi-source parameters to determine the switching timing corresponding to the target driving mode. Thus, by introducing a switching timing prediction mechanism based on distance and speed calculations, the problem of mode mismatch after entering the target road segment caused by reaction lag in traditional driving mode switching is solved. Finally, based on the target driving mode and switching timing, mode switching commands are sent to multiple control units of the vehicle to control the vehicle to switch to the target driving mode. Thus, by distributing commands based on precise switching timing, early intervention and smooth execution of driving mode switching are achieved, effectively avoiding vehicle handling discomfort or jerking caused by mode switching lag.

[0057] This application solves the technical problem of poor system safety and stability caused by single parameters and unreasonable switching timing in existing driving mode switching schemes. It achieves the goal of planning the switching timing in advance, using multi-parameter dynamic weight fusion to accurately match the driving mode, and executing the switching through smooth transition control. This improves the accuracy and safety of mode switching, eliminates the switching jolt, and optimizes driving comfort.

[0058] Optionally, a weighted fusion analysis is performed on the multi-source parameters based on multiple evaluation dimensions to obtain the fusion analysis results, including:

[0059] Step S121: Preprocess the multi-source parameters to obtain a preprocessed parameter set.

[0060] Step S122: Based on the preprocessed parameter set, determine the current driving scenario type of the vehicle.

[0061] Step S123: Determine the scenario correction coefficients corresponding to multiple evaluation dimensions based on the driving scenario type.

[0062] Step S124: Calculate the dimension weights corresponding to the multiple evaluation dimensions using the scene correction coefficient and the basic weights corresponding to the multiple evaluation dimensions.

[0063] Step S125: The preprocessed parameter set is weighted and accumulated using dimensional weights to obtain the fusion analysis results.

[0064] In this embodiment, the driving mode switching system first preprocesses the multi-source parameters to obtain a preprocessed parameter set. The preprocessed parameter set refers to the standardized data set obtained after numerical verification, outlier removal, and dimensional normalization of the originally collected multi-source parameters. The preprocessing process eliminates dimensional differences and noise interference between different parameters.

[0065] For example, the driving mode switching system can normalize raw data such as vehicle speed, gradient, and road surface adhesion coefficient to obtain a preprocessed parameter set.

[0066] Therefore, validity verification eliminates invalid data caused by sensor malfunctions or signal interference, preventing erroneous data from leading to decision-making errors.

[0067] Secondly, the driving mode switching system determines the current driving scenario type of the vehicle based on a preprocessed parameter set.

[0068] Driving scenario type refers to the typical working condition category of the vehicle currently in, determined based on the pre-processed parameter characteristics (such as slope, curvature, adhesion coefficient, etc.), such as icy and slippery scenarios, long downhill scenarios, highway straight-line scenarios, or urban congestion scenarios.

[0069] For example, the driving mode switching system can determine that the current scenario is a long downhill slope based on the detected large negative gradient.

[0070] Therefore, by identifying the specific driving scenario, a basis is provided for the subsequent dynamic adjustment of parameter weights.

[0071] Then, the driving mode switching system determines the scene correction coefficients corresponding to multiple evaluation dimensions based on the driving scenario type.

[0072] The scenario correction factor refers to the correction factor used to dynamically adjust the importance of each evaluation dimension in a specific driving scenario, thereby reflecting the degree of emphasis on dimensions such as safety, terrain, energy consumption or user habits in different scenarios.

[0073] For example, the driving mode switching system can determine the terrain correction factor as 0.4, the safety correction factor as 0.2, and the others as 0 based on the scenario.

[0074] Therefore, by matching the corresponding scene correction coefficient according to the identified scene type, the system can adaptively increase the weight of key dimensions and reduce the weight of non-key dimensions in the current scene.

[0075] Next, the driving mode switching system uses the scene correction coefficient and the basic weights corresponding to multiple evaluation dimensions to calculate the dimension weights corresponding to each of the multiple evaluation dimensions.

[0076] Dimension weight refers to the final weight value calculated by combining the basic weight and the scenario correction coefficient, which is used to characterize the actual influence of each evaluation dimension in the current scenario.

[0077] For example, the driving mode switching system calculates the terrain dimension weights: for instance, the original base weight is 12% + 10%, which is corrected to (12% + 10%). (1+0.4)=30.8%.

[0078] Thus, by combining basic weights with scenario correction coefficients, the fusion of static prior knowledge and dynamic real-time information is achieved.

[0079] Finally, the driving mode switching system uses dimensional weights to perform weighted summation on the preprocessed parameter set to obtain the fusion analysis results.

[0080] For example, the driving mode switching system multiplies the normalized slope, curvature and other parameters by the corresponding dimensional weights and sums them to obtain the total fusion value for terrain, while also calculating the total fusion value for safety, and finally outputs the fusion analysis results.

[0081] Thus, by using weighted summation, multi-dimensional, dispersed information is integrated into a representative total fusion value, achieving a comprehensive representation of multi-source information.

[0082] Optionally, the multi-source parameters are preprocessed to obtain a preprocessed parameter set including:

[0083] Step S1211: Perform validity verification on the multi-source parameters and obtain the verification results.

[0084] Step S1212: When the verification results show that all multi-source parameters are valid, the multi-source parameters are normalized to obtain a preprocessed parameter set.

[0085] Step S1213: When the verification result shows that the navigation parameters are invalid, the real-time terrain parameters in the state-aware parameters are used to replace the navigation parameters to obtain the preprocessed parameter set.

[0086] Step S1214: When the verification result shows that the road surface adhesion coefficient in the state perception parameters is invalid, at least one of the environmental early warning information, road surface friction state information and environmental temperature information in the state perception parameters is used to replace the road surface adhesion coefficient to obtain the preprocessed parameter set.

[0087] Step S1215: When the verification result shows that a single sensor parameter in the state perception parameters is invalid, cross-validation is performed using other environmental parameters in the state perception parameters other than the single sensor parameter to obtain the verification result, and the verification result is used to replace the single sensor parameter to obtain the preprocessed parameter set.

[0088] In this embodiment, the driving mode switching system first performs validity verification on the multi-source parameters and obtains the verification results. Validity verification refers to the system checking the numerical range, signal integrity, and logical rationality of the collected multi-source parameters (including navigation parameters, state perception parameters, etc.) to determine whether the data is usable.

[0089] As can be seen, the driving mode switching system first verifies the validity of all collected multi-source parameters, checking whether the parameters are within a reasonable range (such as vehicle speed 0-200km / h, adhesion coefficient 0.1-0.8) and whether the signal is continuous, and then generates the verification result.

[0090] This enables timely identification and labeling of invalid or abnormal data, preventing erroneous data caused by sensor malfunctions or signal interference from entering the decision-making process. This ensures the quality of subsequent data processing from the source and improves the reliability of the system.

[0091] Secondly, when the verification results show that all multi-source parameters are valid, the driving mode switching system normalizes the multi-source parameters to obtain a preprocessed parameter set.

[0092] As can be seen, if the verification results show that all parameters are valid, the driving mode switching system will directly normalize all parameters to eliminate dimensional differences and generate a preprocessed parameter set.

[0093] This ensures that parameters with different dimensions and ranges can be weighted within a unified numerical range, eliminating the influence of dimensions on weight allocation and providing a standardized data foundation for subsequent accurate weight fusion.

[0094] Then, when the verification result indicates that the navigation parameters are invalid, the driving mode switching system uses the real-time terrain parameters in the state perception parameters to replace the navigation parameters, thus obtaining a preprocessed parameter set.

[0095] Among them, navigation parameter failure refers to the state in which the navigation system is unable to provide information on the road conditions ahead (such as distance, road type, etc.) due to reasons such as signal obstruction (such as tunnels) or equipment failure.

[0096] State perception parameters refer to the vehicle's operating status and surrounding environment information collected in real time by onboard sensors, such as wheel speed, acceleration, camera recognition results, and rain sensor data.

[0097] As can be seen, if the verification result shows that the navigation parameters are invalid, the system will switch to using real-time terrain parameters (such as the current curvature of the curve and the trend of the slope) to infer the characteristics of the road ahead, use these real-time parameters to replace the missing navigation prediction information, and combine them with other effective parameters to generate a preprocessed parameter set.

[0098] This allows the system to maintain basic mode switching functions by judging road segment characteristics in real time, even in areas where high-precision maps are unavailable, thus avoiding system paralysis caused by the failure of a single information source.

[0099] Next, the verification results show that when the road surface adhesion coefficient in the state perception parameters fails, the driving mode switching system uses at least one of the environmental warning information, road surface friction state information and ambient temperature information in the state perception parameters to replace the road surface adhesion coefficient, and obtains a preprocessed parameter set.

[0100] As can be seen, if the verification result shows that the road surface adhesion coefficient is invalid, the system uses other environmental parameters (such as snow and ice warning, road surface friction state identified by the camera, and ambient temperature) to make logical judgments. If a snow and ice warning or low temperature and slippery road surface are detected, a lower safe adhesion coefficient is assigned by default, and this replacement value is used to generate a preprocessing parameter set.

[0101] Therefore, by using conservative default values, the system prioritizes security and prevents misjudgments of patterns due to missing parameters.

[0102] Finally, when the verification result indicates that a single sensor parameter in the state perception parameters has failed, the driving mode switching system uses other environmental parameters in the state perception parameters other than the single sensor parameter to perform cross-validation, obtains the verification result, and uses the verification result to replace the single sensor parameter to obtain the preprocessed parameter set.

[0103] As can be seen, if the verification result shows that a single sensor (such as a rain sensor) is malfunctioning, the system uses other available environmental parameters (such as camera recognition of water accumulation and air humidity) for cross-verification. If the verification result indicates a rainy day, the corresponding rainy day parameter value is assigned, and this verification result is used to replace the data of the malfunctioning sensor to generate a preprocessed parameter set.

[0104] When the main parameter fails, this application employs a strategy to generate equivalent replacement parameters by using other relevant or redundant parameters through specific logical deduction or cross-validation, including real-time terrain parameter replacement, environmental information replacement, and cross-validation replacement.

[0105] Thus, the credibility of alternative data is improved by cross-verifying information from multiple sources, thereby enhancing the system's survivability and decision-making accuracy under partial hardware failures.

[0106] For example, assuming the rain sensor malfunctions, the system checks and finds that the camera has detected water accumulation on the road surface and the air humidity is greater than 80%. Cross-validation determines it to be a rainy day, and the rainfall parameter is assigned a "high" level value to replace the faulty data.

[0107] Optionally, based on the preprocessed parameter set, the driving scenario type is determined as follows:

[0108] Step S1221: Extract driving scenario determination parameters from the preprocessed parameter set. The driving scenario determination parameters include at least one of the following: environmental warning information, road surface adhesion coefficient, long slope identification, slope information, road type information, and vehicle speed information.

[0109] Step S1222: Match the driving scenario determination parameters with the preset scenario determination rules to determine the driving scenario type. The driving scenario type includes at least one of the following: icy and slippery scenario, long downhill and steep slope scenario, highway straight road scenario, urban congestion scenario, and ordinary scenario.

[0110] In this embodiment of the application, firstly, the driving mode switching system extracts driving scenario determination parameters from the preprocessed parameter set. The driving scenario determination parameters refer to the key feature indicators selected from the preprocessed multi-source parameters and used to identify the current macroscopic driving conditions of the vehicle.

[0111] It can be seen that the driving mode switching system extracts core parameters directly related to scene recognition, namely driving scene determination parameters, such as road surface adhesion coefficient. Slope S, road type R, and vehicle speed V, etc.

[0112] For example, long slope identifiers are extracted from the preprocessed parameter set. =1 (or the absolute value of the gradient S > 0.75) and the current vehicle speed =0.67 (corresponding to 80km / h).

[0113] Therefore, by accurately extracting key parameters related to scene recognition from the full set of preprocessed parameters, data redundancy in subsequent logical judgments is reduced, and the computational efficiency of scene recognition is improved.

[0114] Then, the driving mode switching system uses driving scenario determination parameters to match with preset scenario determination rules to determine the driving scenario type. The driving scenario type refers to the typical driving condition state of the vehicle at present, which is determined after matching according to preset rules.

[0115] The preset scenario judgment rules refer to the set of logical judgment conditions pre-stored within the system. They define the specific driving scenario categories corresponding to different parameter combinations or threshold ranges. These can be divided into ice and snow slippery scenarios, long downhill and steep slope scenarios, highway straight road scenarios, urban congestion scenarios, and ordinary scenarios. Each scenario represents a specific driving challenge and demand priority.

[0116] As can be seen, the driving mode switching system inputs the extracted driving scenario determination parameters into the preset scenario determination rule library for comparison and matching. Based on whether the parameter values ​​meet specific thresholds or combinations of conditions, it logically determines which driving scenario the vehicle is currently in. For example, it determines whether the characteristic conditions of ice and snow or slippery conditions are met, or whether the characteristic conditions of long downhill slopes are met, thereby determining the final driving scenario type.

[0117] For example, according to the preset rule "if or If the condition is met, it is determined to be a long downhill steep slope scenario. The extracted parameters are matched with the rules to confirm that the current vehicle is in a long downhill steep slope scenario, and this scenario type is used as the basis for subsequent weight correction.

[0118] This enables the system to clearly distinguish different driving conditions, providing a clear classification basis for dynamically adjusting parameter weights for different scenarios and ensuring the scenario adaptability of mode switching decisions.

[0119] Optionally, multiple evaluation dimensions include safety evaluation, terrain evaluation, energy consumption evaluation, and user habit evaluation. The safety evaluation dimension refers to a set of parameters primarily reflecting vehicle driving safety and road surface adhesion, used to assess the threat level to vehicle stability under current operating conditions, and is the highest priority consideration for mode switching. The terrain evaluation dimension refers to a set of parameters reflecting road geometry and complexity, used to assess the vehicle's driving path morphology (such as slope and curvature) to adapt to the vehicle's handling performance requirements.

[0120] Energy consumption assessment dimensions refer to a set of parameters that reflect a vehicle's energy consumption level and power economy, used to optimize a vehicle's fuel economy or electrical efficiency while meeting safety and terrain requirements.

[0121] User habit assessment dimensions refer to a set of parameters that reflect a driver's personal driving preferences and historical behavior patterns. These parameters are used to personalize the driving experience and balance the system's automatic decision-making with the user's subjective intentions.

[0122] Based on the driving scenario type, the scenario correction coefficients corresponding to multiple evaluation dimensions are determined as follows:

[0123] Step S1231: When the driving scenario type is an ice and snowy slippery scenario, the scenario correction coefficient of the safety assessment dimension is determined as the first value, and the scenario correction coefficients of the terrain assessment dimension, energy consumption assessment dimension and user habit assessment dimension are determined as the second value, wherein the first value is greater than the second value.

[0124] Step S1232: When the driving scenario type is a long downhill and steep slope scenario, the scenario correction coefficient of the terrain assessment dimension is determined as the third value, and the scenario correction coefficients of the safety assessment dimension, energy consumption assessment dimension and user habit assessment dimension are determined as the fourth value, wherein the third value is greater than the fourth value.

[0125] Step S1233: When the driving scenario type is a high-speed straight road scenario, the scenario correction coefficients for the energy consumption assessment dimension and the user habit assessment dimension are determined as the fifth value, and the scenario correction coefficients for the safety assessment dimension and the terrain assessment dimension are determined as the sixth value, wherein the fifth value is greater than the sixth value.

[0126] Step S1234: When the driving scenario type is urban congestion scenario, the scenario correction coefficient of the user habit assessment dimension is determined as the seventh value, and the scenario correction coefficients of the safety assessment dimension, terrain assessment dimension and energy consumption assessment dimension are determined as the eighth value, wherein the seventh value is greater than the eighth value.

[0127] Step S1235: When the driving scenario type is a normal scenario, the scenario correction coefficients corresponding to multiple evaluation dimensions are determined as the ninth value.

[0128] In this embodiment of the application, when the driving scenario is an ice and snowy and slippery scenario, the driving mode switching system determines the scenario correction coefficient of the safety assessment dimension as the first value, and determines the scenario correction coefficients of the terrain assessment dimension, energy consumption assessment dimension and user habit assessment dimension as the second value, wherein the first value is greater than the second value.

[0129] The first through ninth values ​​are preset correction coefficients used to quantify the priority differences of each evaluation dimension in different scenarios. For example, the first value represents a high-priority correction value for the safety dimension in a snow and ice scenario, the second value represents a low-priority correction value for other dimensions in the same scenario, the third value represents a high-priority correction value for the terrain dimension in a long downhill scenario, and so on. The magnitude of these values ​​reflects the degree of emphasis the system places on a particular type of evaluation dimension in a specific scenario.

[0130] It can be seen that, for icy and slippery scenarios, the driving mode switching system sets the correction coefficient of the safety assessment dimension, which represents the importance of safety, to a higher first value, while setting the correction coefficients of the terrain, energy consumption, and user habits dimensions to a lower second value, in order to highlight the safety requirements under icy and snowy road conditions.

[0131] For example, when the system determines that the scenario is icy and slippery, the correction coefficient for the safety assessment dimension... Set to 0.5 (first value), a correction factor for terrain, energy consumption, and user habits. α is set to 0.1 (the second value). This means that in icy and snowy scenarios, the influence of safety parameters (such as road surface adhesion coefficient) on the final mode decision is significantly amplified, while the influence of terrain parameters is reduced.

[0132] Therefore, in icy and slippery scenarios, by significantly increasing the correction coefficient of the safety assessment dimension, the system is forced to prioritize the road surface adhesion and anti-skid requirements, ensuring that the automatic switching mode can maximize driving safety and prevent the risk of ignoring low-adhesion road surfaces due to excessive pursuit of comfort or energy saving.

[0133] When the driving scenario is a long downhill or steep slope scenario, the driving mode switching system determines the scenario correction coefficient of the terrain assessment dimension as the third value, and determines the scenario correction coefficients of the safety assessment dimension, energy consumption assessment dimension, and user habit assessment dimension as the fourth value, wherein the third value is greater than the fourth value.

[0134] It can be seen that, when dealing with long downhill and steep slope scenarios, the driving mode switching system sets the correction coefficient of the terrain assessment dimension, which represents the complexity of road conditions, to a higher third value, and sets the other dimensions to a lower fourth value, in order to highlight the needs of vehicle handling and braking on long downhill slopes.

[0135] For example, when the system determines that the scene is a long downhill steep slope, the correction coefficient for the terrain assessment dimension is... α is set to 0.4 (the third value), a correction factor for safety, energy consumption, and user habits. α is set to 0.2 (the fourth value). This indicates that when descending long slopes, terrain parameters such as slope and curvature become the main basis for decision-making, and the system tends to select modes that can adapt to complex terrain, such as comfort mode to utilize engine braking or optimize energy recovery.

[0136] Therefore, in long downhill and steep slope scenarios, by increasing the correction coefficient of the terrain assessment dimension, the system focuses on road geometric features such as slope and curvature, thereby selecting a mode that is more suitable for coping with the effects of gravity and braking load, and optimizing the vehicle's handling stability and braking performance during the downhill process.

[0137] When the driving scenario is a high-speed straight road scenario, the driving mode switching system determines the scenario correction coefficients for the energy consumption assessment dimension and the user habit assessment dimension as the fifth value, and determines the scenario correction coefficients for the safety assessment dimension and the terrain assessment dimension as the sixth value. The fifth value is greater than the sixth value.

[0138] It can be seen that, for high-speed straight road scenarios, the correction coefficients for the energy consumption assessment dimension and the user habit assessment dimension, which represent economy and personal preference, are set to a higher fifth value, while the correction coefficients for the safety and terrain dimensions are set to a lower sixth value, because the safety of high-speed straight roads is relatively basic, and the focus is on energy saving and comfort / sports preferences.

[0139] For example, when the system determines it to be a high-speed straight-line scenario, the correction coefficients for energy consumption and user habits are... α is set to 0.3 and 0.2 (fifth numerical correlation), correction factors for safety and terrain dimensions. α is set to 0 (sixth value). At this point, the system pays more attention to fuel economy or whether the user prefers sport mode, because the safety risks are relatively low on high-speed straight roads and the terrain is flat.

[0140] Therefore, in high-speed straight-road scenarios, by increasing the correction coefficients for energy consumption and user habits, the system is guided to consider fuel economy or the driver's personal preferences more while ensuring basic safety, thereby achieving a balance between energy efficiency and experience and avoiding reducing driving pleasure due to excessive conservatism in simple road conditions.

[0141] When the driving scenario is an urban congestion scenario, the driving mode switching system determines the scenario correction coefficient for the user habit assessment dimension as the seventh value, and the scenario correction coefficients for the safety assessment dimension, terrain assessment dimension, and energy consumption assessment dimension as the eighth value. The seventh value is greater than the eighth value.

[0142] It can be seen that, for urban congestion scenarios, the correction coefficient for the user habit evaluation dimension is set to a relatively high seventh value, while other dimensions are set to a relatively low eighth value, because in congested road sections, users are particularly sensitive to their personal preferences for start-stop smoothness and kinetic energy recovery intensity.

[0143] Therefore, in urban congestion scenarios, by increasing the correction coefficient of the user habit assessment dimension, respecting the driver's personalized preferences when frequently starting and stopping and following other vehicles at low speeds, driving comfort can be improved and frequent manual intervention caused by conflicts between automatic mode and user habits can be reduced.

[0144] When the driving scenario type is normal, the driving mode switching system determines the scenario correction coefficient corresponding to multiple evaluation dimensions as the ninth value.

[0145] As can be seen, for ordinary scenarios, setting the correction coefficients of all dimensions to the ninth value (usually 0 or the baseline value) means that in the absence of obvious special working conditions, the weights of each dimension maintain the initial proportions and are not dynamically tilted.

[0146] For example, when the system fails to recognize the above-mentioned special scenario, the correction coefficients for all dimensions are... α is set to 0 (the ninth value). At this point, the weights of each dimension are determined solely by the base weights, without any additional bias.

[0147] Therefore, in normal scenarios, all correction coefficients are set to a uniform benchmark value, which avoids introducing unnecessary weight disturbances when there are no special working conditions, maintains the simplicity and stability of the decision-making logic, ensures that the system operates smoothly according to the preset basic logic under normal road conditions, and prevents decision oscillations caused by overly complex dynamic adjustments.

[0148] Optionally, based on the fusion analysis results, the target driving mode is determined from multiple candidate driving modes, including:

[0149] Step S131: Based on the numerical range corresponding to the fusion analysis results, map the fusion analysis results into fuzzy linguistic variables.

[0150] Step S132: Use preset pattern reasoning rules to reason about fuzzy linguistic variables and determine the membership degree corresponding to various candidate driving modes.

[0151] Step S133: Select the target driving mode from multiple candidate driving modes according to the degree of membership.

[0152] In this embodiment, the driving mode switching system first maps the fusion analysis results into fuzzy linguistic variables based on the numerical range corresponding to the fusion analysis results. Fuzzy linguistic variables refer to converting quantitative fusion analysis results (such as total safety fusion values, total terrain fusion values, etc.) into qualitative descriptive terms with semantic meaning, such as "low," "medium," "high," or "smooth," "complex," etc., to simulate the fuzziness of human driving experience.

[0153] As can be seen, the driving mode switching system maps these continuous values ​​in the fusion analysis results to corresponding fuzzy linguistic variables according to the preset numerical range division standard. For example, if the total fusion value of the safety category is in the range of 0.7-1.0, it is mapped to "high".

[0154] Thus, by transforming precise but discrete numerical indicators into linguistic variables that align with human intuition, the system can more flexibly handle fused data in critical states, thereby improving the robustness of decision-making.

[0155] Then, the driving mode switching system uses preset pattern inference rules to reason about fuzzy linguistic variables and determine the membership degrees corresponding to various candidate driving modes. The preset pattern inference rules can be a rule base built on "if-then" logic, used to establish the mapping relationship between fuzzy linguistic variables and candidate driving modes.

[0156] It can be seen that the driving mode switching system calls the preset mode reasoning rule library, takes the fuzzy linguistic variables output by the previous steps as input, and calculates the membership degree of each candidate driving mode (such as snow mode, comfort mode, etc.) under the current working conditions through fuzzy logic operations (such as intersection, union, etc.).

[0157] This enables a logical mapping from multi-dimensional fuzzy linguistic variables to specific driving modes, comprehensively considering the interactive effects of multiple factors such as safety, terrain, and energy consumption. As a result, a quantitative degree of matching (membership) is given for each candidate mode, providing a scientific basis for the final selection.

[0158] Finally, the driving mode switching system selects the target driving mode from multiple candidate driving modes based on membership degree. Membership degree refers to a numerical index calculated according to fuzzy inference rules, indicating how well a candidate driving mode matches the current vehicle's driving state. The value typically ranges from 0 to 1; a higher value indicates that the mode is more suitable for the current operating conditions.

[0159] As can be seen, the driving mode switching system compares the membership values ​​of all candidate driving modes and selects the mode with the highest membership value as the target driving mode.

[0160] This ensures that the selected driving mode is the most compatible with the current operating conditions, thereby guaranteeing the rationality and accuracy of mode switching and effectively improving the vehicle's adaptability and driving experience.

[0161] Optionally, the fusion analysis results include fusion values ​​for safety assessment, terrain assessment, and energy consumption assessment dimensions; multiple candidate driving modes include snow mode, comfort mode, economy mode, and sport mode.

[0162] Among them, the safety assessment dimension fusion value refers to the numerical index that reflects the current road safety and anti-skid requirements after weighting safety parameters such as road surface adhesion coefficient and snow and ice warning.

[0163] The terrain assessment dimension fusion value refers to a numerical indicator that reflects the complexity and difficulty of driving a road, obtained by weighting terrain parameters such as curve curvature, slope, and long slope markings.

[0164] The energy consumption assessment dimension fusion value refers to a numerical indicator that reflects the current energy consumption level and economic needs after weighting and calculation based on energy consumption parameters such as road type and vehicle speed.

[0165] Among the various candidate driving modes, Snow Mode is designed for low-friction surfaces and prioritizes maximizing safety; Comfort Mode is designed for complex terrain or general road conditions and balances handling stability and ride smoothness; Eco Mode is designed for efficient driving and prioritizes optimizing fuel economy or energy efficiency; and Sport Mode is designed for high-performance driving and prioritizes providing agile power response and precise handling.

[0166] Using pattern inference rules to reason about fuzzy linguistic variables, the membership degrees corresponding to various candidate driving modes are determined, including:

[0167] Step S1321: Map the fused values ​​of the safety assessment dimension to safety level linguistic variables, map the fused values ​​of the terrain assessment dimension to terrain level linguistic variables, and map the fused values ​​of the energy consumption assessment dimension to energy consumption level linguistic variables.

[0168] Step S1322: Based on the pattern reasoning rules, reason about the safety level linguistic variables, terrain level linguistic variables, energy consumption level linguistic variables, and user habit parameters to determine the membership degrees corresponding to snow mode, comfort mode, economy mode, and sports mode, respectively.

[0169] In this embodiment of the application, firstly, the driving mode switching system maps the safety assessment dimension fusion value to a safety level linguistic variable, the terrain assessment dimension fusion value to a terrain level linguistic variable, and the energy consumption assessment dimension fusion value to an energy consumption level linguistic variable.

[0170] Among them, the safety level linguistic variable refers to the qualitative descriptive terms obtained by fuzzy mapping the fused values ​​of the safety assessment dimensions, which are used to characterize the safety risk level of the current working condition, such as "low", "medium", and "high".

[0171] The terrain level linguistic variable refers to the qualitative descriptive terms obtained by fuzzing the fused values ​​of terrain assessment dimensions, which are used to characterize the complexity of road terrain, such as "smooth" or "complex".

[0172] Energy consumption level linguistic variables refer to qualitative descriptive terms obtained by fuzzy mapping of the fused values ​​of energy consumption assessment dimensions. They are used to characterize the degree of energy consumption demand under the current operating conditions, such as "low demand", "medium demand", and "high demand".

[0173] As can be seen, the driving mode switching system receives the quantitative fusion analysis results from the preceding steps, namely the fusion values ​​of the safety assessment dimension, the terrain assessment dimension, and the energy consumption assessment dimension. Subsequently, based on the preset numerical interval mapping relationship, these three consecutive values ​​are converted into corresponding qualitative linguistic variables: the safety assessment dimension fusion value is converted into a safety level linguistic variable, the terrain assessment dimension fusion value is converted into a terrain level linguistic variable, and the energy consumption assessment dimension fusion value is converted into an energy consumption level linguistic variable.

[0174] This allows the system to respond more flexibly to fused data in a critical state, improving the robustness of decision-making.

[0175] Then, the driving mode switching system uses the mode reasoning rules to infer the safety level linguistic variables, terrain level linguistic variables, energy consumption level linguistic variables, and user habit parameters to determine the membership degree corresponding to snow mode, comfort mode, economy mode, and sport mode.

[0176] As can be seen, the driving mode switching system calls the preset mode reasoning rule base, taking the three levels of linguistic variables output from the previous steps and user habit parameters as input conditions for logical reasoning. By matching the rules in the rule base, it calculates the membership values ​​corresponding to the four candidate driving modes—snow mode, comfort mode, economy mode, and sport mode—thereby quantifying the degree of adaptability of each mode to the current operating conditions.

[0177] Therefore, by combining multiple dimensions of fuzzy linguistic variables with user habits through pattern reasoning rules, the membership degree of each candidate driving mode can be comprehensively derived.

[0178] Optionally, navigation prediction analysis can be performed using multi-source parameters to determine the switching timing, including:

[0179] Step S141: Extract the predicted distance corresponding to the target road segment from the navigation parameters.

[0180] Step S142: Extract the vehicle's current real-time speed from the state perception parameters.

[0181] Step S143: Determine the switching timing based on the predicted distance, real-time vehicle speed, and preset mode smooth transition duration.

[0182] In this embodiment of the application, firstly, the driving mode switching system extracts the predicted distance corresponding to the target road segment from the navigation parameters.

[0183] Among them, the predicted distance is the straight-line distance or path distance between the vehicle's current position and the starting point of the target road segment (such as snow, long downhill, etc.) provided in the navigation parameters.

[0184] As can be seen, the driving mode switching system extracts the predicted distance corresponding to the target road segment from the navigation parameters in the multi-source parameters. This distance represents how far the vehicle is from reaching the specific road conditions (such as icy and snowy roads or long downhill slopes) where the mode needs to be switched.

[0185] Therefore, by extracting the predicted distance, the system can quantify the spatial location of the target road conditions, providing a spatial dimension basis for advance planning of mode switching, enabling the system to have predictive capabilities rather than passive responses.

[0186] Then, the driving mode switching system extracts the vehicle's current real-time speed from the state perception parameters.

[0187] As can be seen, the driving mode switching system extracts the vehicle's current real-time speed from the state perception parameters, which represents how quickly the vehicle approaches the target road segment.

[0188] Therefore, by extracting real-time vehicle speed, the system can grasp the vehicle's time-dimensional motion state and calculate the time window to reach the target road conditions by combining distance information, ensuring that the calculation of the switching timing is in line with the actual driving rhythm of the vehicle.

[0189] Finally, the driving mode switching system determines the switching timing based on the predicted distance, real-time vehicle speed, and preset mode transition duration. The mode transition duration refers to the preset time required for the system to adjust parameters such as power, suspension, and steering from the old mode to the new mode to eliminate any potential jerking during mode switching.

[0190] As can be seen, the driving mode switching system substitutes the predicted distance, real-time vehicle speed, and the system's preset mode smooth transition duration into the switching timing calculation formula to calculate how far in advance the system should start the switching process, thereby determining the specific switching timing.

[0191] Thus, by comprehensively predicting distance, real-time vehicle speed, and smooth transition duration to determine the switching timing, precise synchronization between mode switching actions and changes in road conditions is achieved.

[0192] Optionally, depending on the target driving mode and the timing of the switch, mode switching commands may be sent to multiple control units, including:

[0193] Step S151: Based on the target driving mode and the switching timing, determine the parameter adjustment information corresponding to multiple control units, including: powertrain control unit, body control system, suspension control system and steering system.

[0194] Step S152: Based on the parameter adjustment information, send mode switching commands to multiple control units respectively to control the multiple control units to perform parameter adjustment actions in a linear gradual manner.

[0195] In this embodiment, firstly, the driving mode switching system determines the parameter adjustment information corresponding to multiple control units based on the target driving mode and the switching timing.

[0196] Among them, parameter adjustment information refers to the control variables and their target values ​​that need to be modified for each control unit (such as powertrain, body control, etc.), such as power response delay time, suspension stiffness, steering assist, etc.

[0197] As can be seen, the driving mode switching system queries a preset mode parameter mapping table based on the determined target driving mode and the switching timing, and generates corresponding specific parameter adjustment information for each control unit.

[0198] Therefore, by determining specific parameter adjustment information for each control unit, the executability and accuracy of mode switching commands are ensured, enabling different subsystems to work together to form the vehicle dynamic characteristics required for the target driving mode.

[0199] Secondly, the driving mode switching system sends mode switching commands to multiple control units based on parameter adjustment information, so as to control multiple control units to perform parameter adjustment actions in a linear and gradual manner.

[0200] Among them, the linear gradual change method refers to the parameter changing uniformly and continuously over time from the current value to the target value, rather than abruptly, in order to eliminate the mechanical shock or sense of power interruption during the switching process.

[0201] As can be seen, the driving mode switching system sends mode switching commands to the four control units mentioned above based on the parameter adjustment information. The commands include the target parameter values ​​and the adjustment time constraints, requiring each control unit to gradually adjust its control parameters in a linear and gradual manner within a preset smooth transition time until the target value is reached.

[0202] Therefore, by sending commands and adjusting linearly, the common problem of parameter abrupt changes during mode switching is effectively solved.

[0203] In summary, this application solves the technical problem of poor system safety and stability caused by single parameters and unreasonable switching timing in existing driving mode switching schemes. It achieves the goal of planning the switching timing in advance, accurately matching the driving mode by using multi-parameter dynamic weight fusion, and executing the switching through smooth transition control. This improves the accuracy and safety of mode switching, eliminates the feeling of switching jerkiness, and optimizes driving comfort.

[0204] This application provides a driving mode switching method that solves the problems of manual switching in existing passenger vehicle driving mode switching, such as lag, misjudgment, and high operational threshold; initial automatic switching parameters being singular, lacking predictive ability, and having weak anti-interference; as well as switching jerks and unscientific timing. This results in the following beneficial effects: improved driving safety, reducing safety risks caused by untimely switching; elimination of switching jerks, optimizing driving comfort; improved mode adaptation accuracy, enhancing system stability; balancing energy consumption and power performance; and lowering the user's operating threshold, adapting to different driving habits.

[0205] Figure 2 This is a process for an automatic switching technology route for passenger vehicle driving modes based on navigation prediction and multi-parameter fusion, according to an embodiment of this application. Figure 2 After the system is started, the information acquisition layer is responsible for synchronously acquiring navigation prediction data (such as curvature C, slope S, and distance D) and real-time perception data (such as adhesion coefficient μ and vehicle speed V). After validity verification, if all parameters are valid, multi-parameter preprocessing is performed, including normalization and outlier removal. When some parameters fail, a fault-tolerant mechanism is used to replace the faulty parameters. Then the data flows to the multi-parameter fusion decision layer, which dynamically adjusts the weights through scene recognition and uses fuzzy logic reasoning combined with user preferences to determine the optimal driving mode. Then the timing of early switching is calculated to eliminate jerking. Next, the system enters the mode execution layer, where it issues instructions to the Electronic Control Unit (ECU) and Body Control Module (BCM), etc., and adjusts parameters such as power and suspension in a linear and gradual manner within the smooth mode switching time to achieve a smooth switch. Finally, the system enters the feedback adjustment layer, which monitors the vehicle's stability and comfort indicators in real time. If the indicators are not met, the decision weights are corrected or the optimal mode is switched. If the indicators are met, the current mode is maintained and parameters are continuously collected.

[0206] Specifically, firstly, after the vehicle is powered on, the automatic driving mode switching system starts simultaneously and completes the following initialization operations:

[0207] Initialization includes loading preset parameters. Specifically, preset parameters include, but are not limited to: a basic weight matrix (as shown in Table 1, which represents the basic weight matrix), parameter thresholds (e.g., which may include the road surface adhesion coefficient), and other parameters. Danger threshold 0.3, high curvature curve threshold 0.05 rad / m), fuzzy decision rule base, mode switching smoothing time. (For example, it can be 1 second), minimum switching interval (for example, it can be arbitrarily preset to 3 seconds).

[0208] Initialization operations may also include initializing multiple functional modules of the vehicle. Specifically, these multiple functional modules may include: navigation system (such as high-precision maps or real-time navigation), vehicle sensor network (including wheel speed sensors, acceleration sensors, suspension sensors, etc.), environmental sensors (such as cameras, millimeter-wave radar, rain sensors, etc.), BCM, and ECU.

[0209] Initialization operations may also include reading user historical data. For example, the automatic driving mode switching system loads the user's driving mode preference records for the past 30 days (such as the frequency of use of Sport mode, trigger scenarios for Eco mode, etc.) for subsequent decision adjustments.

[0210] Table 1

[0211]

[0212] In Table 1, among the parameter types, safety accounts for 40% of the total weight, terrain accounts for 30%, energy consumption accounts for 20%, and user habits account for 10%.

[0213] Among the parameter thresholds, under the safety category, the road surface adhesion coefficient... 25% A value less than 0.3 indicates danger; It performs moderately well between 0.3 and 0.6; A value greater than 0.6 indicates safety; a warning parameter W of 1 for ice / snow or slippery conditions indicates a warning, while W of 0 indicates no warning. In the terrain category, curve curvature C accounts for 12%, with C greater than 0.05 rad / m indicating large curvature and C less than 0.05 rad / m indicating small curvature; slope accounts for 10%, divided into downhill (0° to 12°) and uphill (-12° to 0°); uphill / downhill indicators L account for 8% of the base weight, with L of 1 indicating a long slope and L of 0 indicating a normal road section. In the energy consumption category, road type R accounts for 12% of the base weight, R can be highway, urban, rural, or mountainous; vehicle speed V accounts for 8% of the base weight, divided into V greater than 100 km / h, V between 60 km / h and 100 km / h, and V < 60 km / h. In the user habit category, historical preference H accounts for 10% of the base weight.

[0214] After the system starts up, in order to ensure real-time decision-making, two types of parameters are collected synchronously in real time at a sampling frequency of 10Hz: navigation information and real-time sensing parameters.

[0215] During the navigation information collection process, predictive and static parameters are combined. Specifically, the vehicle navigation system analyzes road segment data for a distance of 1 to 3 kilometers ahead and extracts the following information as navigation information from the road segment data: road segment type (R represents highway, urban / rural, or mountainous), road surface attributes (e.g., paved or unpaved), curve curvature (C), gradient (S), long slope indicator (L=1 indicates the current road segment is a long slope, L=0 indicates the current road segment is a normal road segment), environmental warning (W=1 indicates the current road segment is an icy or slippery road segment, W=0 indicates the current road segment is not an icy or slippery road segment), predicted distance (denoted as D, representing the straight-line distance between the target road segment ahead and the current vehicle, such as 500 meters or 1 kilometer), and geographical location (including latitude and longitude, to determine whether it is a high-latitude snow-prone area).

[0216] Specifically, if the vehicle's automatic driving mode switching system detects that the navigation signal strength is weak (such as when the vehicle is in a tunnel), it records a navigation signal failure indicator and collects navigation information again after the navigation signal is restored.

[0217] During real-time sensing parameter acquisition, vehicle state parameters and environmental parameters are combined. Specifically, road surface parameters are calculated using wheel speed sensors and accelerometers to determine the road adhesion coefficient. (The calculation formula is) Where F_x is the longitudinal force of the tire, m is the vehicle's curb weight, and a_x is the longitudinal acceleration; the road surface friction condition (dry, wet, snowy, or icy road conditions) is identified by a camera; the tire slip ratio is read by the Electronic Stability Program (ESP) system;

[0218] Specifically, the vehicle status parameter system can read the real-time vehicle speed V (in km / h, converted to m / s for subsequent calculations) through the vehicle speed sensor; read the brake pressure through the brake pedal sensor; read the steering angle through the steering angle sensor; read the suspension compression through the suspension displacement sensor; and read the current vehicle load through the load sensor.

[0219] During the acquisition of environmental parameters, the system can read real-time rainfall through a rain sensor (for example, to determine whether it is a rainy day); read ambient temperature through a temperature sensor (to enhance the weight of snow and ice warnings when the temperature is below 0℃); and read visibility through a visibility sensor.

[0220] After the information acquisition layer, parameter validity verification and fault tolerance processing are performed. First, the validity of all collected parameters is verified to avoid decision-making errors caused by invalid data. The verification rules include numerical range verification. For example, the reasonable range for the road surface adhesion coefficient μ is 0.1 to 0.8. If it exceeds this range, it is determined to be invalid. The reasonable range for vehicle speed V is 0 to 200 km / h. If it exceeds this range, it is invalid.

[0221] Secondly, all collected parameters undergo signal integrity verification. If a parameter fails to return data for three consecutive collection cycles (e.g., navigation signal loss), it is considered a parameter failure. If some parameters fail, a fault-tolerant mechanism (e.g., redundancy replacement) is implemented for the collected parameters. If navigation parameters fail (e.g., D is 0), real-time terrain parameters are used to replace the predicted information, and the continuous changes in curve curvature C and slope S are used to determine whether it is a mountainous or long-slope road section. If the road surface adhesion coefficient μ fails, it is replaced by the combination of ice and snow warning W and road surface friction state with ambient temperature. If W is 1 or the ambient temperature is less than 0℃ and the road surface friction state is "snow or ice," then μ is defaulted to 0.2 (hazard level). If a single transmission... If a sensor parameter fails (e.g., a rain sensor malfunctions), other environmental parameters are used for cross-validation. For example, if the camera detects water accumulation on the road surface and the air humidity is greater than 80%, it is determined to be a rainy day. If at least three of the core parameters related to safety, terrain, and energy consumption fail simultaneously (e.g., navigation, wheel speed sensors, and the camera all fail), the system automatically downgrades to "Comfort Mode" and displays a message on the instrument panel stating "Automatic driving mode switching is limited; please switch manually," while maintaining basic driving functions. If all parameters are valid, multi-parameter preprocessing is performed, namely normalization combined with outlier removal. Valid parameters that pass verification are preprocessed to eliminate dimensional differences and abnormal fluctuations. This includes outlier removal using… The principle is to eliminate random outliers and calculate the mean μ and standard deviation of a certain parameter. If the parameter value exceeds If the range is too wide, it is considered an outlier and replaced with the parameter value from the previous period. Normalization is also performed: parameters of different dimensions are uniformly mapped to the [0,1] interval to avoid the influence of dimensions on weight calculation. The formula is: Where x is the original parameter value, x_min is the minimum calibrated value of the parameter, and x_max is the maximum calibrated value (e.g., road adhesion coefficient). Speed Output the preprocessed parameter set: (The ' indicates the normalized value).

[0222] Then, multi-parameter fusion decision-making is performed (e.g., weight calculation combined with fuzzy decision-making). First, the final weight is calculated by combining the basic weight with dynamic correction, and then the optimal driving mode is output by fuzzy decision-making. The dynamic weight calculation rules can be referred to in Table 2.

[0223] Table 2

[0224]

[0225] When performing dynamic weight calculations, the system needs to determine the current driving scenario and the correction coefficients for each parameter. (The value selection rules are shown in Table 2, which contains scene correction coefficients.) (The rules for determining the values). The final weight is calculated using the following formula: ,in Based on weights, This refers to the scene correction coefficients. The method for calculating the fusion value of various parameters is as follows: the total fusion value of a certain type of parameter is Σ (the normalized value of each parameter in that type × the final weight), for example, the total fusion value for the security category. .

[0226] When the system performs fuzzy decision-making (i.e., pattern matching), it needs to fuzzify the multi-source fusion indicators. Specifically, the system maps the numerical ranges of the total fusion value S_total (security category), T_total (terrain category), and E_total (energy consumption category) to corresponding fuzzy linguistic variables. Specifically, S_total is mapped to three fuzzy linguistic variables: low (0-0.3), medium (0.3-0.7), and high (0.7-1.0); T_total is mapped to two fuzzy linguistic variables: smooth (0-0.4) and complex (0.4-1.0); and E_total is divided into three fuzzy linguistic variables: low demand (0-0.3), medium demand (0.3-0.7), and high demand (0.7-1.0).

[0227] After the system makes a fuzzy decision on multiple parameters, it calls the preset "if-then" rule base to determine the optimal driving mode. The rule base contains multiple rules that can cover a variety of driving scenarios. For example, multiple rules can be represented as: "If S_total = High → Then Optimal Mode = Snow Mode"; "If S_total = Medium and T_total = Complex → Then Optimal Mode = Comfort Mode"; "If S_total = Medium and T_total = Smooth and E_total = High Demand → Then Optimal Mode = Economy Mode"; "If S_total = Low and T_total = Smooth and E_total = Low Demand and H' > 0.6 (User Prefers Sport) → Then Optimal Mode = Sport Mode"; "If S_total = Low and T_total = Complex and H' > 0.6 (User Prefers Comfort) → Then Optimal Mode = Comfort Mode"; "If S_total = Medium and T_total = Smooth and E_total = Medium Demand → Then Optimal Mode = User's Historical High-Frequency Mode".

[0228] In addition, based on the actual needs of the application scenario, the following rules can be set: "If S_total is medium, T_total is flat, and E_total is medium demand, the system determines it as the user's historical high-frequency mode"; "If S_total is low, T_total is complex, and E_total is high demand, the system determines it as the comfort mode".

[0229] After determining the optimal driving mode, the system defuzzifies and calculates the dynamic switching time.

[0230] In the defuzzing step, the "centroid method" is used to calculate the membership degree of each mode (e.g., snow mode has a membership degree of 0.9, comfort mode has a membership degree of 0.1), and the mode with the highest membership degree is the "optimal driving mode"; if the difference in membership degree between two modes is less than or equal to 0.05 (e.g., comfort mode has a membership degree of 0.52, economy mode has a membership degree of 0.48), then the user's historical preference is used ( Based on the data, select the mode that users use more frequently.

[0231] In the dynamic mode switching timing calculation step, to avoid jerking or handling inaccuracies caused by switching modes only after entering the target road segment, the system calculates the "advance switching time" based on the navigation-predicted distance and current vehicle speed, ensuring that the mode switch is completed before entering the road segment. The calculation formula is as follows: Where D represents the distance to the target road segment ahead as predicted by the navigation (in meters, such as 500 meters); V represents the current real-time speed of the vehicle (in meters per second, converted from km / h, such as 60 km / h = 16.7 m / s). Indicates the smooth transition time of the mode (the calibrated value is 0.5-2 seconds, and the default value is 1 second to ensure that the parameter adjustment is smooth).

[0232] For example, if the road ahead is 500 meters away from snow (D is 500m), and the current vehicle speed is 60km / h (V is 16.7m / s). If the time to switch in advance is 1 second, then the advance switching time t = 500 / 16.7 + 1 ≈ 31 seconds; the system starts the switching process when it is 500 meters away from the snowy road section, and the switching is completed after 31 seconds, that is, the switching is completed just before entering the snowy road section.

[0233] In special scenario adjustments, if the predicted distance D is less than 100 meters (short-distance emergency scenario), then Adjusted to 0.5 seconds to speed up the switching process; if the vehicle speed V is less than 30km / h (low-speed scenario), the calculation formula is as follows: To avoid slow switching, use seconds.

[0234] Following the dynamic switching timing calculation step is the mode execution layer, which is used to smoothly switch to the optimal mode.

[0235] The system sends mode switching commands to each control unit, achieving a smooth transition through "gradual parameter adjustment" to avoid any jerking.

[0236] The system first sends switching commands to the ECU, BCM, suspension control unit, and steering system, specifying the parameter adjustment targets for each system (as shown in Table 3, which outlines the core adjustment targets for the four driving modes in the example); subsequently, each system adjusts its parameters in a "linear gradual" manner, with a mode switching smoothness time of [time missing]. (e.g., 1 second) - For example, when switching from Eco mode to Sport mode, the power response speed is linearly adjusted from "0.5 seconds delay" to "no delay", and reaches the standard parameters of Sport mode after 1 second; then the instrument panel displays "Switching to XX mode soon" (before switching) and "Switched to XX mode" (after switching), and at the same time, it informs the user of the current mode status through voice prompts (e.g., "Snowy road ahead, switched to Snow mode"), and retains the user's manual intervention right (the user can force the switch mode through the steering wheel buttons, and the system records the operation and uses it for subsequent preference correction).

[0237] Table 3

[0238]

[0239] Table 4 lists the control objects as Economy Mode, Comfort Mode, Sport Mode, and Junior Mode, and sets the power response, shift logic, suspension stiffness, steering assist, energy recovery, and traction control according to the above four modes.

[0240] Following the model execution layer is the feedback adjustment layer, used for real-time monitoring and closed-loop correction. After the mode switch is completed, the system continuously monitors the vehicle status, dynamically corrects the decision logic, forms closed-loop control, and collects vehicle stability parameters (such as lateral acceleration, longitudinal slip ratio, and body vibration amplitude), ride comfort parameters (such as suspension vibration frequency and power output fluctuation), and energy consumption parameters (such as fuel consumption or electricity consumption per 100 kilometers) in real time after the switch.

[0241] For example, the performance standards include standards for stability, comfort, and energy consumption. The stability standards include a lateral acceleration of less than or equal to 0.8g and a longitudinal slip ratio of less than or equal to 5% (or 8% or less in snow mode); the comfort standards include a body vibration amplitude of less than or equal to 0.1g and a suspension vibration frequency of less than or equal to 5Hz; and the energy consumption standards include fuel consumption or electricity consumption in economy mode that is less than or equal to the average level of vehicles in the same class.

[0242] If the condition meets the standard, the current mode is maintained, parameters are collected again, and the next decision-making cycle begins. If the condition does not meet the standard, for example, if the longitudinal slip rate is 10% in snow mode (not meeting the standard), the weight of safety parameters is automatically adjusted (the weight of the road adhesion coefficient μ is increased to 40%), and the power output is further limited (the maximum torque is reduced by another 10%) until the condition meets the standard.

[0243] Finally, the system performs self-learning corrections and monitors user manual intervention behaviors over a long period of time (such as users switching from economy mode to sport mode multiple times in a highway straight-line scenario). The system automatically adjusts the weights of energy consumption and user habits in this scenario, such as reducing the weight of energy consumption by 5% and increasing the weight of user habits by 5%, and then prioritizing the triggering of sport mode in subsequent similar scenarios.

[0244] According to a specific embodiment of this application (taking a long downhill scenario as an example), a pure electric passenger vehicle is used as the application carrier. The vehicle is equipped with a high-precision navigation system, wheel speed sensor, acceleration sensor, slope sensor, suspension displacement sensor and body control unit. The specific implementation steps are as follows.

[0245] Step 1: System initialization, loading basic weight matrix (safety category 40%, terrain category 30%, energy consumption category 20%, user habit category 10%), slope calibration range [-12°, 12°], and mode switching smoothing time. = 1 second, and read the user's historical preferences (e.g., high-frequency mode to comfort mode).

[0246] Step 2: Multi-source parameter acquisition. The navigation system analyzes that the road ahead is a long downhill section of 1 km (L' is 1), and predicts the distance D to be 1000 m. Real-time collected parameters show that the vehicle speed V is 80 km / h (converted to 22.2 m / s), the slope S is -7° (S' is 0.875 after normalization), and the road surface adhesion coefficient is... The value is 0.65 (μ' is 0.786 after normalization), the braking pressure is 0.2 MPa, and the sampling frequency is 10 Hz.

[0247] Step 3: Data preprocessing and verification. The system uses 3D processing. Outliers are removed in principle, and all parameters are normalized to ensure that all parameters are valid after verification and there are no invalid items.

[0248] Step 4: Scene recognition and dynamic weight calculation. The system determines whether the scene is a long downhill slope or a steep slope. Specifically, for safety-related scenarios... 0.2, terrain type 0.4, energy consumption category α = 0, User Habits α is 0; the final weights are calculated as follows: security category 48%, terrain category 42%, energy consumption category 20%, and user habit category 10%; after fusion, S_total is 0.65, T_total is 0.85, and E_total is 0.3.

[0249] Step 5: Fuzzy decision-making. Specifically, S_total represents medium, T_total represents complex, and E_total represents low demand. The system combines user preferences to determine that the optimal mode is the comfort mode.

[0250] Step 6: Calculate the switching timing using the formula. The second-by-second calculation system initiates the switching process when the distance to the long downhill section is 1000 meters.

[0251] Step 7: Mode execution. The system sends instructions to the BCM and ECU, and linearly and gradually adjusts the parameters (transition time 1 second) to a power response delay of 0.15 seconds, softens the suspension, and adjusts energy recovery to level 2. The instrument panel displays "Switching to comfort mode is imminent" and provides a voice prompt after the switch is complete.

[0252] Step 8: Feedback correction, real-time monitoring of lateral acceleration (0.5g) and vehicle vibration amplitude (0.08g), both of which meet the standards; continuously collect parameters and enter the next cycle to ensure that the comfort mode is maintained throughout the long downhill section, reducing braking load.

[0253] In this embodiment of the application, the road surface adhesion coefficient The calibration range is [0.1, 0.8], and the corresponding normalization formula is: The vehicle speed V calibration range is: The normalization formula is The slope S calibration range is: The normalization formula is The aforementioned calibration range can be flexibly adjusted according to different passenger vehicle models. This application's embodiments are adaptable to various types of passenger vehicles, including gasoline, pure electric, and hybrid vehicles. Specifically, for gasoline vehicles, the focus is on adjusting the transmission shifting logic; for pure electric vehicles, the focus is on adjusting the energy recovery intensity; and for hybrid vehicles, the control logic of both gasoline and hybrid electric vehicles can be combined. When applying the method provided in this application's embodiments to the aforementioned passenger vehicles, no new hardware is required; it can be implemented through software upgrades.

[0254] In addition, if the navigation signal is lost (D is 0), the real-time slope S and curve curvature C are used to replace the predicted information to determine the long downhill scene, and the mode switching can still be triggered normally to ensure the stable operation of the system.

[0255] In summary, the embodiments of this application achieve relevant beneficial effects through the following technical points.

[0256] First, this application proposes a collaborative decision-making logic that combines navigation prediction and multi-parameter fusion. By combining navigation prediction distance with real-time parameters, it enables early switching of driving modes, overcoming the limitations of passive response in existing solutions.

[0257] Secondly, this application proposes a dynamic weight allocation mechanism for multi-dimensional parameters, which dynamically adjusts the weight of each parameter according to the priority of the scenario to ensure the accuracy of mode switching under different working conditions.

[0258] Then, the embodiments of this application propose a smooth transition control for mode switching and a scientific calculation of the switching timing. By adjusting parameters linearly and gradually, and combining vehicle speed and predicted distance to calculate the switching time, the sense of jerkiness is eliminated.

[0259] Next, this application proposes a multi-parameter redundant acquisition and fault-tolerant replacement mechanism to improve the system's anti-interference capability and avoid decision-making errors caused by the failure of a single sensor.

[0260] Finally, this application proposes a closed-loop correction and self-learning optimization logic, which dynamically corrects decisions by monitoring the vehicle status in real time to adapt to different users' driving habits.

[0261] In particular, in some other application scenarios, special implementation methods can be used to achieve the technical solutions of the embodiments of this application.

[0262] For example, if the in-vehicle high-precision map is unavailable, the road type and terrain markings of ordinary real-time navigation can be used, combined with the vehicle's real-time terrain parameters (curvature of curves, slope) to make navigation predictions, thereby realizing the function of switching driving modes in advance.

[0263] For example, dynamic weight calculation can use a neural network algorithm instead of a weighted fusion algorithm, and fuzzy decision-making can use a decision tree algorithm instead. As long as accurate fusion and pattern matching of multiple parameters can be achieved, they are all within the protection scope of this application.

[0264] For example, linear gradual adjustment can be replaced with segmented gradual adjustment, and the transition time can be dynamically adjusted according to vehicle speed and load, thereby eliminating the jerky feeling when switching driving modes.

[0265] According to an embodiment of this application, a device embodiment for controlling a driving mode switching system is provided. It should be noted that the device can be used to execute the above-described method for controlling the driving mode switching system.

[0266] Figure 3 This is a structural block diagram of a driving mode switching device according to one embodiment of this application, such as... Figure 3As shown, taking the control device 300 of the driving mode switching system as an example, the device includes: an acquisition module 301 for acquiring multi-source parameters of the vehicle, wherein the parameter types of the multi-source parameters include navigation parameters and state perception parameters; an analysis module 302 for performing weighted fusion analysis on the multi-source parameters based on multiple evaluation dimensions to obtain fusion analysis results, wherein the fusion analysis results are used to characterize the fusion features of the vehicle in the current driving environment; a first determination module 303 for determining the target driving mode from multiple candidate driving modes according to the fusion analysis results; a second determination module 304 for performing navigation prediction analysis using multi-source parameters to determine the switching timing corresponding to the target driving mode; and a control module 305 for sending mode switching commands to multiple control units of the vehicle according to the target driving mode and the switching timing to control the vehicle to switch to the target driving mode.

[0267] Embodiments of this application also provide a vehicle, including: a memory storing an executable program; and a processor for running the program, wherein the program executes the methods described in various embodiments of this application when it runs.

[0268] Embodiments of this application also provide a computer-readable storage medium including a stored executable program, wherein, when the executable program is running, it controls the device where the computer-readable storage medium is located to perform the methods of various embodiments of this application.

[0269] Embodiments of this application also provide a computer program product, including a computer program that, when executed by a processor, implements the methods of various embodiments of this application.

[0270] Embodiments of this application also provide a computer program product, including a non-volatile computer-readable storage medium for storing a computer program that, when executed by a processor, implements the methods in various embodiments of this application.

[0271] Embodiments of this application also provide a computer program that, when executed by a processor, implements the methods described in the various embodiments of this application.

[0272] In the above embodiments of this application, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0273] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units can be a logical functional division, and in actual implementation, there may be other division methods. For instance, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling, direct coupling, or communication connection may be through some interfaces; the indirect coupling or communication connection between units or modules may be electrical or other forms.

[0274] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0275] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0276] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, read-only memory (ROM), random access memory (RAM), portable hard drive, magnetic disk, or optical disk.

[0277] The above description is only a preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A method for switching driving modes, characterized in that, include: Acquire multi-source parameters of the vehicle, wherein the parameter types of the multi-source parameters include navigation parameters and state perception parameters; The multi-source parameters are weighted and fused based on multiple evaluation dimensions to obtain fusion analysis results, wherein the fusion analysis results are used to characterize the fusion features of the vehicle in the current driving environment; Based on the fusion analysis results, the target driving mode is determined from multiple candidate driving modes; The navigation prediction analysis is performed using the multi-source parameters to determine the switching timing corresponding to the target driving mode. Based on the target driving mode and the switching timing, a mode switching command is sent to multiple control units of the vehicle to control the vehicle to switch to the target driving mode.

2. The method according to claim 1, characterized in that, The weighted fusion analysis of the multi-source parameters based on the multiple evaluation dimensions yields the following results: The multi-source parameters are preprocessed to obtain a preprocessed parameter set; Based on the preprocessed parameter set, the current driving scenario type of the vehicle is determined; Based on the driving scenario type, determine the scenario correction coefficients corresponding to the multiple evaluation dimensions respectively; Using the scenario correction coefficient and the basic weights corresponding to the multiple evaluation dimensions, calculate the dimension weights corresponding to the multiple evaluation dimensions. The preprocessed parameter set is weighted and accumulated using the dimensional weights to obtain the fusion analysis result.

3. The method according to claim 2, characterized in that, The multi-source parameters are preprocessed to obtain the preprocessed parameter set, which includes: The validity of the multi-source parameters is verified to obtain the verification results; When the verification result shows that all the multi-source parameters are valid, the multi-source parameters are normalized to obtain the preprocessed parameter set. When the verification result indicates that the navigation parameters are invalid, the real-time terrain parameters in the state-aware parameters are used to replace the navigation parameters to obtain the preprocessed parameter set; When the verification result indicates that the road surface adhesion coefficient in the state perception parameters fails, at least one of the environmental early warning information, road surface friction state information and environmental temperature information in the state perception parameters is used to replace the road surface adhesion coefficient to obtain the preprocessing parameter set. When the verification result indicates that a single sensor parameter in the state perception parameters is invalid, cross-validation is performed using other environmental parameters in the state perception parameters other than the single sensor parameter to obtain the verification result, and the verification result is used to replace the single sensor parameter to obtain the preprocessed parameter set.

4. The method according to claim 2, characterized in that, Based on the preprocessed parameter set, the driving scenario type is determined as follows: Driving scenario determination parameters are extracted from the preprocessed parameter set, wherein the driving scenario determination parameters include at least one of the following: environmental warning information, road surface adhesion coefficient, long slope identification, slope information, road type information, and vehicle speed information; The driving scenario type is determined by matching the driving scenario determination parameters with preset scenario determination rules. The driving scenario type includes at least one of the following: icy and slippery scenario, long downhill and steep slope scenario, highway straight road scenario, urban congestion scenario, and ordinary scenario.

5. The method according to claim 2, characterized in that, The multiple evaluation dimensions include safety evaluation, terrain evaluation, energy consumption evaluation, and user habit evaluation; based on the driving scenario type, the scenario correction coefficients corresponding to the multiple evaluation dimensions are determined as follows: When the driving scenario type is an icy and slippery scenario, the scenario correction coefficient of the safety assessment dimension is determined as a first value, and the scenario correction coefficients of the terrain assessment dimension, the energy consumption assessment dimension and the user habit assessment dimension are determined as a second value, wherein the first value is greater than the second value. When the driving scenario type is a long downhill and steep slope scenario, the scenario correction coefficient of the terrain assessment dimension is determined as the third value, and the scenario correction coefficients of the safety assessment dimension, the energy consumption assessment dimension and the user habit assessment dimension are determined as the fourth value, wherein the third value is greater than the fourth value. When the driving scenario type is a high-speed straight road scenario, the scenario correction coefficients of the energy consumption assessment dimension and the user habit assessment dimension are determined as the fifth value, and the scenario correction coefficients of the safety assessment dimension and the terrain assessment dimension are determined as the sixth value, wherein the fifth value is greater than the sixth value. When the driving scenario type is an urban congestion scenario, the scenario correction coefficient of the user habit assessment dimension is determined as the seventh value, and the scenario correction coefficients of the safety assessment dimension, the terrain assessment dimension and the energy consumption assessment dimension are determined as the eighth value, wherein the seventh value is greater than the eighth value. When the driving scenario type is a normal scenario, the scenario correction coefficient corresponding to the multiple evaluation dimensions is determined as the ninth value.

6. The method according to claim 1, characterized in that, Based on the fusion analysis results, the target driving mode is determined from the multiple candidate driving modes, including: Based on the numerical range corresponding to the fusion analysis results, the fusion analysis results are mapped to fuzzy linguistic variables; The fuzzy linguistic variables are inferred using preset pattern inference rules to determine the membership degrees corresponding to the various candidate driving modes. The target driving mode is selected from the multiple candidate driving modes according to the membership degree.

7. The method according to claim 6, characterized in that, The fusion analysis results include fusion values ​​for safety assessment, terrain assessment, and energy consumption assessment dimensions; the multiple candidate driving modes include snow mode, comfort mode, economy mode, and sport mode; the fuzzy linguistic variables are inferred using the mode inference rules to determine the membership degrees corresponding to the multiple candidate driving modes, including: The fused values ​​of the security assessment dimensions are mapped to security level linguistic variables, the fused values ​​of the terrain assessment dimensions are mapped to terrain level linguistic variables, and the fused values ​​of the energy consumption assessment dimensions are mapped to energy consumption level linguistic variables. Based on the pattern reasoning rules, reasoning is performed on the safety level linguistic variables, terrain level linguistic variables, energy consumption level linguistic variables, and user habit parameters to determine the membership degrees corresponding to the snow mode, comfort mode, economy mode, and sports mode, respectively.

8. The method according to claim 1, characterized in that, Using the multi-source parameters for navigation prediction analysis to determine the switching timing includes: Extract the predicted distance corresponding to the target road segment from the navigation parameters; Extract the vehicle's current real-time speed from the state perception parameters; The switching timing is determined based on the predicted distance, the real-time vehicle speed, and the preset mode smooth transition duration.

9. The method according to claim 1, characterized in that, Sending mode switching commands to the plurality of control units based on the target driving mode and the switching timing includes: Based on the target driving mode and the switching timing, parameter adjustment information corresponding to the plurality of control units is determined, wherein the plurality of control units include: a powertrain control device, a body control system, a suspension control system, and a steering system; Based on the parameter adjustment information, mode switching commands are sent to the multiple control units respectively to control the multiple control units to perform parameter adjustment actions in a linear gradual manner.

10. A vehicle, characterized in that, It includes a memory and a processor, wherein the memory stores a computer program and the processor is configured to run the computer program to perform the driving mode switching method of any one of claims 1 to 9.

Images