A new energy vehicle cooperative control fault processing method and system

Abstract

The application belongs to the technical field of new energy vehicle control, and discloses a new energy vehicle cooperative control fault processing method and system, the method comprising: acquiring vehicle operation data, environment perception data, driver operation data and driver attention related data; according to the acquired data, evaluating the driving scene complexity and the driver cognitive load; receiving the fault information output by the cooperative control system; according to the fault information, the driving scene complexity and the driver cognitive load, selecting an information presentation strategy from a preset strategy library; the information presentation strategy comprises an information presentation channel, information content detail and a presentation time; according to the selected information presentation strategy, generating fault information, and controlling the human-computer interface device to execute information presentation; the fault information can be conveyed to the driver in a clear, timely and non-interfering driving safety manner according to the driving scene complexity and the driver cognitive load, so as to avoid safety risks or damage to the driving experience caused by improper information transmission.

Description

Technical Field

[0001] This application relates to the field of new energy vehicle control technology, and more specifically, to a method and system for handling faults in the collaborative control of new energy vehicles. Background Technology

[0002] The collaborative control system of new energy vehicles is crucial for their safe and stable operation. It coordinates the work of multiple core subsystems, including the electric drive system, braking system, and energy management system, to achieve the driver's intentions or preset driving goals. During vehicle operation, any subsystem within the collaborative control system may malfunction. To address this, collaborative control systems typically incorporate sophisticated fault diagnosis functions. When the system detects abnormal fluctuations in data from a subsystem, it initiates an internal fault diagnosis process, using a series of algorithmic analyses and logical judgments to accurately identify the specific fault type.

[0003] Once a fault is diagnosed, the cooperative control system further assesses its potential impact on vehicle performance and driving safety. Based on the assessment results, the system immediately takes appropriate fault-handling measures, such as limiting the vehicle's maximum power output, adjusting the torque distribution of other normally operating drive motors, disabling some non-critical functions, and even entering a specific performance-limiting "limp" mode depending on the severity of the fault, to ensure that the vehicle can continue to drive or stop relatively safely after the fault occurs. While the system takes these fault-handling measures, a crucial aspect is the timely and accurate communication of fault information and the resulting changes in vehicle status to the driver; this is an indispensable part of ensuring driving safety.

[0004] However, existing technologies have significant shortcomings in presenting fault information. Traditional fault information presentation methods are often too simplistic and generic, typically limited to illuminating a general fault indicator light on the dashboard or displaying a vague text warning on the central control screen, such as "Powertrain Fault" or "Please Check Vehicle." This simple and generic approach fails to meet the needs of complex driving scenarios. Upon receiving such ambiguous warnings, drivers often cannot clearly understand the specific nature of the fault (e.g., is it a motor or battery fault?), its severity (e.g., is it a minor fault or an emergency requiring immediate stopping?), its specific impact on vehicle performance (e.g., how much power has been reduced? Has braking performance been affected?), and the system's already taken or about-to-be-taken countermeasures. This lack of transparency easily leads to confusion, uncertainty, and even anxiety among drivers.

[0005] Furthermore, if a malfunction occurs while the driver is traveling at high speed, preparing to change lanes or overtake, or at a critical moment requiring high concentration and precise operation, a sudden, vague warning message from the system, possibly accompanied by an audible warning, can severely distract the driver. Due to the unclear information received, the driver may become anxious and unable to make optimal judgments, or even take panicked and inappropriate emergency actions, such as sudden, sharp braking or steering, which actually increases the risk of an accident. Even if the fault handling measures taken by the cooperative control system are smooth at the vehicle dynamics level, the psychological stress response of the driver due to insufficient information will still significantly affect their perception of the vehicle's status and the accuracy of subsequent actions, thereby reducing the overall driving experience and safety.

[0006] Therefore, in the process of new energy vehicle collaborative control system failure and response measures, how to overcome the limitations of existing technologies in fault information presentation, especially in complex driving scenarios, and how to clearly, timely and non-disruptively communicate fault information, vehicle status changes and system response measures to drivers and passengers through an intelligent and adaptive information presentation and human-computer interaction strategy, thereby reducing driver anxiety and uncertainty and avoiding safety risks or damage to the driving experience caused by improper information transmission, has become an urgent technical problem to be solved in the field of human-computer interaction design of new energy vehicle collaborative control system.

[0007] To address the aforementioned issues, existing technologies urgently need improvement. Summary of the Invention

[0008] The purpose of this application is to provide a method and system for handling faults in collaborative control of new energy vehicles, which can convey fault information to the driver in a clear, timely and non-disruptive manner according to the complexity of the driving scenario and the driver's cognitive load, thereby reducing the driver's anxiety and uncertainty, and avoiding safety risks or damage to the driving experience caused by improper information transmission.

[0009] Firstly, this application provides a method for handling faults in the collaborative control system of a new energy vehicle, used to present fault information to the driver when a fault occurs in the collaborative control system of a new energy vehicle during vehicle operation. The method includes the following steps:

[0010] A1. Acquire vehicle operation data, environmental perception data, driver operation data, and driver attention-related data;

[0011] A2. Based on the acquired vehicle operation data, environmental perception data, driver operation data, and driver attention-related data, assess the complexity of the driving scenario and the driver's cognitive load;

[0012] A3. Receive fault information output by the cooperative control system; the fault information includes fault type, fault severity, and vehicle performance impact assessment results;

[0013] A4. Based on fault information, the complexity of the driving scenario, and the driver's cognitive load, select an information presentation strategy from the preset strategy library; the information presentation strategy includes the information presentation channel, the level of detail of the information content, and the timing of presentation.

[0014] A5. Based on the selected information presentation strategy, generate fault information and control the human-machine interface device to perform information presentation.

[0015] Preferably, step A1 includes:

[0016] A101. Read vehicle operation data via the vehicle bus CAN interface; vehicle operation data includes vehicle speed, steering angle, and acceleration;

[0017] A102. Acquire environmental perception data through radar and cameras; environmental perception data includes distance to surrounding vehicles, relative speed, and road curvature.

[0018] A103. Acquire driver operation data through sensors; driver operation data includes throttle opening, brake pressure, and steering torque;

[0019] A104. Obtain driver attention-related data through image recognition; driver attention-related data includes eye movement data and head posture.

[0020] A105. Preprocess vehicle operation data, environmental perception data, driver operation data, and driver attention-related data.

[0021] Preferably, step A2 includes:

[0022] A201. Based on vehicle operation data, assess the vehicle's driving stability and obtain a stability index;

[0023] A202. Based on environmental perception data and road type, assess the degree of road congestion to obtain congestion level indicators;

[0024] A203. Based on driver operation data and driver attention-related data, assess the driver's fatigue level and distraction level to obtain fatigue level index and distraction level index;

[0025] A204. Calculate the complexity of driving scenarios and the cognitive load of drivers by combining stability indicators, congestion indicators, fatigue indicators, and distraction indicators.

[0026] Preferably, step A201 includes:

[0027] Based on vehicle operation data, extract vehicle speed sequence, steering angle sequence, and acceleration sequence;

[0028] For the vehicle speed sequence, steering angle sequence, and acceleration sequence, calculate the sequence entropy to obtain the vehicle speed sequence entropy, steering angle sequence entropy, and acceleration sequence entropy.

[0029] Based on the vehicle speed sequence entropy, steering angle sequence entropy, and acceleration sequence entropy, a weighted fusion algorithm is used to calculate the stability index.

[0030] Preferably, step A202 includes:

[0031] The road type of the road where the vehicle is located is determined based on the vehicle's real-time location;

[0032] Based on environmental perception data, extract the distance sequence of surrounding vehicles, the relative speed sequence, and the road curvature sequence;

[0033] For the distance sequence and relative speed sequence of surrounding vehicles, a dynamic time warping algorithm is used to calculate the distance congestion factor and speed congestion factor;

[0034] For the road curvature sequence, calculate the rate of change of curvature, and adjust the distance congestion factor and speed congestion factor according to the road type and the rate of change of curvature.

[0035] Based on the corrected distance congestion factor and the corrected speed congestion factor, a weighted fusion algorithm is used to calculate the road congestion index.

[0036] Preferably, the step of calculating the rate of change of curvature for the road curvature sequence, and correcting the distance congestion factor and speed congestion factor according to the road type and the rate of change of curvature includes:

[0037] The reference curvature change rate threshold is determined based on the road type where the vehicle is located; road types include highways, urban roads and rural roads, and different road types correspond to different reference curvature change rate thresholds.

[0038] For road curvature sequences, a sliding window algorithm is used to calculate the curvature change rate sequence; the length of the sliding window is adaptively adjusted according to the vehicle speed, and the higher the vehicle speed, the longer the window length.

[0039] For the curvature change rate sequence, count the number of curvature change rates that are greater than the baseline curvature change rate threshold to obtain the number of curvature change rates exceeding the threshold.

[0040] Based on the road type and the number of curvature change rates exceeding the threshold, a piecewise function is used to correct the distance congestion factor and the speed congestion factor; the correction range for highways is smaller than that for urban roads and rural roads, and the more curvature change rates exceeding the threshold, the greater the correction range.

[0041] Preferably, step A203 includes:

[0042] Based on driver operation data, throttle opening sequence, brake pressure sequence and steering torque sequence are extracted, and the mean and variance of each sequence are calculated.

[0043] Based on driver attention-related data, blink frequency from eye movement data and head-turning frequency from head posture data were extracted.

[0044] For the mean and variance of the throttle opening sequence, the mean and variance of the braking pressure sequence, the mean and variance of the steering torque sequence, the blink frequency and the nodding frequency, a preset fuzzy inference rule is used to calculate the driver fatigue level index and the driver distraction level index.

[0045] Preferably, step A204 includes:

[0046] The stability index, congestion index, fatigue index, and distraction index were normalized.

[0047] The driving scenario complexity is obtained by weighted fusion calculation of the difference between 1 and the normalized stability index and the normalized congestion index.

[0048] The driver's cognitive load is obtained by weighted fusion calculation of the normalized fatigue level index and the normalized distraction level index.

[0049] Preferably, step A4 includes:

[0050] A401. Prioritize fault information based on fault type, fault severity, and vehicle performance impact assessment results;

[0051] A402. Assess the risk level of a driving scenario based on its complexity;

[0052] A403. Determine the urgency of information presentation based on the priority of fault information and the risk level of the driving scenario;

[0053] A404. Select an information presentation strategy from a preset strategy library based on the urgency of the information presentation and the driver's cognitive load; the information presentation strategy includes the information presentation channel, the level of detail of the information content, and the timing of presentation.

[0054] Secondly, this application provides a fault handling system for collaborative control of new energy vehicles, used to present fault information to the driver when a fault occurs in the collaborative control system of a new energy vehicle during vehicle operation. The system includes:

[0055] The data acquisition module is used to acquire vehicle operation data, environmental perception data, driver operation data, and driver attention-related data.

[0056] The evaluation module is used to assess the complexity of the driving scenario and the cognitive load of the driver based on the acquired vehicle operation data, environmental perception data, driver operation data, and driver attention-related data.

[0057] The fault information receiving module is used to receive fault information output by the cooperative control system; the fault information includes fault type, fault severity and vehicle performance impact assessment results;

[0058] The strategy selection module is used to select an information presentation strategy from a preset strategy library based on fault information, the complexity of the driving scenario, and the driver's cognitive load. The information presentation strategy includes the information presentation channel, the level of detail of the information content, and the timing of presentation.

[0059] The information presentation module is used to generate fault information based on the selected information presentation strategy and control the human-machine interface device to perform information presentation.

[0060] Beneficial effects: The new energy vehicle collaborative control fault handling method and system provided in this application acquires multi-dimensional data and evaluates driving scenarios and driver status. Combined with fault information, it intelligently selects appropriate information presentation strategies, thereby solving the problem that the fault information presentation methods in the prior art are simple and general, which are difficult to meet the needs of complex driving scenarios. It can convey fault information to the driver in a clear, timely and non-disruptive manner according to the complexity of the driving scenario and the driver's cognitive load, thereby reducing the driver's anxiety and uncertainty, and avoiding safety risks or damage to the driving experience caused by improper information transmission. Attached Figure Description

[0061] Figure 1 A flowchart of a new energy vehicle collaborative control fault handling method provided in an embodiment of this application.

[0062] Figure 2 This is a schematic diagram of the structure of the new energy vehicle collaborative control fault handling system provided in the embodiments of this application.

[0063] Labeling Explanation: 1. Data Acquisition Module; 2. Evaluation Module; 3. Fault Information Receiving Module; 4. Strategy Selection Module; 5. Information Presentation Module. Detailed Implementation

[0064] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. The components of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0065] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this application, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0066] refer to Figure 1 This application proposes a fault handling method for the collaborative control system of new energy vehicles, which is used to present fault information to the driver when a fault occurs in the collaborative control system of a new energy vehicle during vehicle operation. The steps of the method include:

[0067] A1. Acquire vehicle operation data, environmental perception data, driver operation data, and driver attention-related data;

[0068] A2. Based on the acquired vehicle operation data, environmental perception data, driver operation data, and driver attention-related data, assess the complexity of the driving scenario and the driver's cognitive load;

[0069] A3. Receive fault information output by the cooperative control system; the fault information includes fault type, fault severity, and vehicle performance impact assessment results;

[0070] A4. Based on fault information, the complexity of the driving scenario, and the driver's cognitive load, select an information presentation strategy from the preset strategy library; the information presentation strategy includes the information presentation channel, the level of detail of the information content, and the timing of presentation.

[0071] A5. Based on the selected information presentation strategy, generate fault information and control the human-machine interface device to perform information presentation.

[0072] Among these, vehicle operation data refers to data reflecting the current motion state of the vehicle, used to assess vehicle driving stability. Environmental perception data refers to data reflecting the state of the environment surrounding the vehicle, used to assess road congestion levels. Driver operation data refers to data reflecting the driver's control operations on the vehicle, used to assess driver fatigue and distraction levels. Driver attention-related data refers to data reflecting the driver's attention state, used to assess driver fatigue and distraction levels.

[0073] Among these, the complexity of the driving scenario refers to the degree to which the current driving environment occupies the driver's attention and the potential risks, used to determine the urgency of information presentation. Driver cognitive load refers to the psychological burden on the driver currently processing information, used to select information presentation strategies.

[0074] Among them, fault information refers to the details of faults diagnosed by the collaborative control system, which is used to assess the priority of fault information.

[0075] The preset strategy library refers to a collection of various information presentation schemes, specifically including different combinations of information presentation channels, information content detail levels, and presentation timings, used to select the appropriate presentation method based on the context. The information presentation strategy refers to the specific plan for conveying fault information to the driver, used to guide the generation and presentation of fault information. The information presentation channel refers to the medium used to transmit fault information, specifically employing one or more methods such as visual, auditory, and tactile, used to select the most effective communication channel based on the context. The information content detail level refers to the level of detail in conveying fault information, specifically employing methods such as concise prompts or detailed explanations, used to adjust the amount of information based on the driver's state and the complexity of the scenario. The presentation timing refers to the point in time when the fault information is displayed or communicated to the driver, specifically employing methods such as immediate prompts or delayed prompts, used to avoid interfering with the driver at critical moments.

[0076] Among them, human-machine interface devices refer to the hardware in the vehicle used to interact with the driver, which can be implemented by one or more of the following: instrument panel, central control screen, speaker, steering wheel vibrator, etc., to execute the selected information presentation strategy.

[0077] The core innovation of this application lies in its ability to adaptively select information presentation strategies from a preset strategy library based on fault information, the complexity of the driving scenario, and the driver's cognitive load. These strategies include the information presentation channel, the level of detail in the information content, and the timing of presentation, thereby enabling intelligent and contextualized presentation of fault information.

[0078] This method acquires vehicle operation data, environmental perception data, driver operation data, and driver attention-related data to provide a foundation for subsequent evaluation and decision-making. This data reflects the vehicle's current dynamic state, external environmental conditions, and the driver's real-time behavior and physiological state.

[0079] Furthermore, based on the acquired multi-dimensional data, the complexity of the current driving scenario and the driver's cognitive load are assessed. The assessment of driving scenario complexity reflects the potential occupation of the driver's attention by the current environment and the difficulty of the driving task; the assessment of driver cognitive load reflects the driver's current information processing ability and state. These assessment results provide key criteria for determining when and how to present information to the driver, aiming to avoid interference when the driver is not suited to receive information.

[0080] Simultaneously, it receives fault information output from the cooperative control system. This fault information includes the fault type, severity, and an assessment of the fault's impact on vehicle performance. This detailed fault information forms the basis for determining the specific content that needs to be conveyed to the driver and the urgency of that information.

[0081] Therefore, based on the received fault information, the assessed complexity of the driving scenario, and the driver's cognitive load, the system selects the most suitable information presentation strategy from a pre-defined strategy library. This strategy library contains multiple information presentation schemes, each defining the information presentation channel (e.g., visual, auditory, tactile), the level of detail of the information content, and the timing of information presentation. The strategy selection process is an adaptive decision based on the current situation, aiming to ensure the effectiveness and safety of information delivery.

[0082] Finally, based on the selected information presentation strategy, specific fault information content is generated, and the human-machine interface (HMI) devices are controlled to present the information. HMI devices may include the instrument panel, central control screen, speakers, steering wheel vibrator, etc. By controlling these devices, the generated fault information is presented to the driver through a selected channel, with varying levels of detail and at appropriate times, thus ensuring the safe and clear communication of fault information. The entire process forms a closed loop, enabling the presentation of fault information to dynamically adapt to vehicle operating status, environmental conditions, and driver state.

[0083] By employing the aforementioned solution, this application addresses the problem that existing technologies present fault information in a vague and unresponsive manner when a new energy vehicle's collaborative control system malfunctions, leading to driver confusion, distraction, and even inappropriate actions, thereby impacting driving safety and experience. This solution achieves adaptive presentation of fault information by comprehensively considering fault characteristics, driving scenarios, and driver state, improving the clarity and effectiveness of information delivery. As a result, drivers can more accurately understand the vehicle's status and the impact of the fault, reducing uncertainty and anxiety. Simultaneously, by selecting appropriate presentation channels, levels of detail, and timing, the interference of fault prompts on driver attention is reduced, avoiding driving risks caused by improper information transmission, and enhancing driving safety and user experience.

[0084] Specifically, step A1 includes:

[0085] A101. Read vehicle operation data via the vehicle bus CAN interface; vehicle operation data includes vehicle speed, steering angle, and acceleration;

[0086] A102. Acquire environmental perception data through radar and cameras; environmental perception data includes distance to surrounding vehicles, relative speed, and road curvature.

[0087] A103. Acquire driver operation data through sensors; driver operation data includes throttle opening, brake pressure, and steering torque;

[0088] A104. Obtain driver attention-related data through image recognition; driver attention-related data includes eye movement data and head posture.

[0089] A105. Preprocess vehicle operation data, environmental perception data, driver operation data, and driver attention-related data.

[0090] The vehicle operating data includes vehicle speed, steering angle, and acceleration. These data reflect the vehicle's dynamic state and provide input for assessing vehicle ride stability.

[0091] Environmental perception data includes distances to surrounding vehicles, relative speeds, and road curvature. This data is used to perceive the vehicle's external environment, providing a basis for assessing road congestion and road type. For example, radar can provide distances and relative speeds to surrounding vehicles, while cameras can identify road markings and calculate road curvature.

[0092] The driver's operational data includes throttle opening, brake pressure, and steering torque. This data directly reflects the driver's control input to the vehicle, providing input for assessing driver fatigue and distraction levels. This data can be measured by sensors integrated into the cooperative control system.

[0093] Driver attention-related data includes eye movement data and head posture. This data reflects the driver's physiological and behavioral state, providing evidence for assessing driver fatigue and distraction levels. For example, eye movement data may include blink frequency and gaze direction, while head posture may include the head's pitch and yaw angles relative to the vehicle's direction. This driver attention-related data can be obtained by capturing images of the driver using in-vehicle cameras and then identifying them using existing image recognition technologies.

[0094] The process involves preprocessing the four types of acquired data. Preprocessing may include data cleaning, such as removing noise and outliers; data synchronization, such as time alignment of data collected from different sensors; and data format conversion, such as converting raw data into a unified format. This ensures the usability of the data input to subsequent evaluation modules.

[0095] The vehicle operation data, environmental perception data, driver operation data, and driver attention-related data acquired through the above steps provide a foundation for subsequent assessments of driving scenario complexity and driver cognitive load. Comprehensive data acquisition and preprocessing ensure that the assessment process is based on high-quality data, thereby supporting the selection of subsequent information presentation strategies and resolving issues of inappropriate information presentation due to insufficient or unavailable data.

[0096] In some implementations, step A2 includes:

[0097] A201. Based on vehicle operation data, assess the vehicle's driving stability and obtain a stability index;

[0098] A202. Based on environmental perception data and road type, assess the degree of road congestion to obtain congestion level indicators;

[0099] A203. Based on driver operation data and driver attention-related data, assess the driver's fatigue level and distraction level to obtain fatigue level index and distraction level index;

[0100] A204. Calculate the complexity of driving scenarios and the cognitive load of drivers by combining stability indicators, congestion indicators, fatigue indicators, and distraction indicators.

[0101] This solution proposes specific assessment steps to address the question of how to evaluate the complexity of driving scenarios and the cognitive load on drivers.

[0102] First, in step A201, the vehicle's driving stability is assessed based on vehicle operation data to obtain a stability index. The vehicle's driving stability reflects the current driving environment and driving operation status. For example, in complex road conditions or during emergency obstacle avoidance, the vehicle's driving will exhibit instability. Assessing stability can reflect the state of the driving scenario.

[0103] Next, in step A202, the degree of road congestion is assessed based on environmental perception data and road type to obtain a congestion level index. Road congestion is a factor affecting the driving scenario state; congested road conditions require drivers to invest more attention and energy. Assessing the congestion level can directly quantify the state of the driving scenario.

[0104] Then, in step A203, the driver's fatigue and distraction levels are assessed based on driver operation data and driver attention-related data to obtain fatigue and distraction indices. The driver's fatigue and distraction status determine their current cognitive load level. Fatigue or distraction reduces the driver's ability to process information and respond to emergencies. Assessing these indicators can reflect the driver's cognitive load.

[0105] Finally, in step A204, the stability index, congestion index, fatigue index, and distraction index obtained from steps A201 to A203 are combined to calculate the complexity of the driving scenario and the driver's cognitive load. This comprehensive evaluation method considers factors from the vehicle, environment, and driver, thus obtaining realistic assessment results of driving scenario complexity and driver cognitive load. These assessment results provide a basis for selecting information presentation strategies based on these factors, enabling the system to adjust the information presentation method according to the actual situation, avoiding distraction of the driver's attention in complex scenarios or under high load conditions, and improving the effectiveness and safety of fault information presentation.

[0106] By applying this assessment method to the fault information presentation process, the system can dynamically adjust the presentation method, content, and timing of fault information based on the driver's current cognitive state and the complexity of the driving environment. This ensures effective information delivery while reducing interference with the driver's attention, avoiding improper operation or safety risks, and solving the problem of inadequate assessment leading to unsuitable information presentation in existing technologies.

[0107] Preferably, step A201 may include:

[0108] Based on vehicle operation data, extract vehicle speed sequence, steering angle sequence, and acceleration sequence;

[0109] For the vehicle speed sequence, steering angle sequence, and acceleration sequence, calculate the sequence entropy to obtain the vehicle speed sequence entropy, steering angle sequence entropy, and acceleration sequence entropy.

[0110] Based on the vehicle speed sequence entropy, steering angle sequence entropy, and acceleration sequence entropy, a weighted fusion algorithm is used to calculate the stability index.

[0111] The process involves extracting vehicle speed, steering angle, and acceleration sequences. The length of these sequences can be set according to actual needs; they can be set to a fixed value or adaptively adjusted based on vehicle speed.

[0112] Furthermore, sequence entropy is an indicator of sequence complexity or uncertainty. Sequence entropy can be calculated using the Shannon entropy algorithm, which requires discretization of the sequence data. Discretization divides a continuous numerical range into several intervals (i.e., "bins"), and then counts the frequency of data points within each interval. The number and width of the bins can be preset or dynamically determined based on the distribution characteristics of the sequence data (the Shannon entropy algorithm is existing technology, and its calculation process will not be detailed here). The calculated vehicle speed sequence entropy, steering angle sequence entropy, and acceleration sequence entropy reflect the smoothness of the vehicle in terms of longitudinal speed change, lateral steering operation, and longitudinal acceleration / deceleration, respectively. Lower sequence entropy values ​​indicate that the corresponding parameters change smoothly, indicating good smoothness; higher sequence entropy values ​​indicate that the corresponding parameters change drastically, indicating poor smoothness.

[0113] Therefore, a weighted fusion algorithm linearly combines the three sequence entropy values, assigning a weight to each entropy value. These weights can be fixed values ​​(e.g., the weight of the vehicle speed sequence entropy can be set to 0.4, the weight of the steering angle sequence entropy to 0.3, and the weight of the acceleration sequence entropy to 0.3), or they can be dynamically adjusted according to the current driving state or scenario (e.g., the weights of vehicle speed and acceleration can be increased when the vehicle is traveling in a straight line; the weight of steering angle can be increased when the vehicle is turning). Through weighted fusion, stationarity information from different dimensions is integrated into a single stationarity index. This stationarity index is used to subsequently assess the complexity of the driving scenario and the driver's cognitive load. By providing a more comprehensive and multi-dimensional quantification of stationarity, this stationarity index provides more refined input for subsequent evaluation steps, thereby helping to improve the accuracy of the assessment of driving scenarios and driver states.

[0114] Preferably, step A202 may include:

[0115] The road type of the road where the vehicle is located is determined based on the vehicle's real-time location;

[0116] Based on environmental perception data, extract the distance sequence of surrounding vehicles, the relative speed sequence, and the road curvature sequence;

[0117] For the distance sequence and relative speed sequence of surrounding vehicles, a dynamic time warping algorithm is used to calculate the distance congestion factor and speed congestion factor;

[0118] For the road curvature sequence, calculate the rate of change of curvature, and adjust the distance congestion factor and speed congestion factor according to the road type and the rate of change of curvature.

[0119] Based on the corrected distance congestion factor and the corrected speed congestion factor, a weighted fusion algorithm is used to calculate the road congestion index.

[0120] This solution provides specific methods for assessing road congestion levels, aiming to more comprehensively and accurately reflect the complexity of actual driving scenarios, thereby providing a more reliable basis for subsequent fault information presentation strategies.

[0121] First, the road type of the vehicle is determined based on its real-time location. This can be achieved by combining the vehicle's GPS data with a pre-installed digital map database. Different types of road environments have a significant impact on the driver's cognitive load and driving complexity; distinguishing road types is the foundation for more refined congestion assessments.

[0122] Next, based on the environmental perception data, the distance sequence to surrounding vehicles, the relative speed sequence, and the road curvature sequence are extracted. These sequence data contain key information reflecting the current traffic conditions (surrounding vehicles) and road geometry (road curvature), and are the basic inputs required to assess the degree of congestion. The length of the extracted sequences can be adjusted according to actual needs.

[0123] Then, for the distance and relative speed sequences of surrounding vehicles, the Dynamic Time Warping (DWT) algorithm is used to calculate the distance congestion factor and speed congestion factor. Using the DWT algorithm to process sequence data can more effectively capture the dynamic characteristics of the changes in distance and relative speed of surrounding vehicles over time, such as the distance and speed change patterns during vehicle platooning or following. This allows for the calculation of distance and speed congestion factors that better reflect vehicle interaction and traffic flow, overcoming the limitations of simple instantaneous or average value evaluations. The DWT algorithm finds the optimal matching path between two time series, calculating their similarity or distance even if they are offset or stretched on the time axis, thus more accurately assessing congestion dynamics based on vehicle spacing and relative speed. The DTW algorithm can be used to calculate the similarity or distance between the current sequence and preset reference sequences representing different congestion states (e.g., free flow, congestion), and map this similarity or distance to a congestion factor. For example, for the surrounding vehicle distance sequence and the relative speed sequence, Euclidean distance is used as a metric, and a dynamic time warping algorithm with Sakoe-Chiba window constraints is applied to calculate the DTW distance between the current sequence and multiple preset urban arterial road congestion reference sequences (reference surrounding vehicle distance sequences or reference relative speed sequences for different congestion levels) representing different levels of congestion. These DTW distances are then mapped to a distance congestion factor or a speed congestion factor (e.g., the sum of the reciprocals of each DTW distance is used as the distance congestion factor or the speed congestion factor).

[0124] Furthermore, for the road curvature sequence, the rate of change of curvature is calculated, and the distance congestion factor and speed congestion factor are corrected based on the road type and the rate of change of curvature. This step is crucial to the scheme, as it incorporates the geometric complexity of the road itself into the congestion assessment. The rate of change of road curvature reflects the degree of road curvature and driving difficulty. The rate of change of curvature can be obtained, for example, by calculating the difference between adjacent points in the curvature sequence or the magnitude of curvature change within a sliding window. By calculating the rate of change of curvature and combining it with the previously determined road type, the congestion factor calculated solely based on vehicle spacing and speed is corrected, ensuring that the congestion assessment considers not only traffic flow but also the characteristics of the road itself and the driving experience under different road types. For example, on road sections with large curvature changes, even with few vehicles, the driving difficulty and "congestion feeling" experienced by drivers may increase; this correction can more accurately reflect this complexity. This correction mechanism allows the congestion index to more closely resemble the actual experience of drivers in specific road environments, thus more accurately reflecting the complexity of driving scenarios. This revised congestion index, along with other indicators such as vehicle ride smoothness, driver fatigue, and distraction, is used to comprehensively calculate the complexity of driving scenarios and the driver's cognitive load. This more accurate congestion assessment, as input, makes the evaluation of the overall complexity of driving scenarios more precise, thereby more effectively guiding the selection of subsequent fault information presentation strategies and ensuring that the most appropriate information is provided to the driver in driving environments of varying complexity.

[0125] Finally, based on the corrected distance congestion factor and the corrected speed congestion factor, a weighted fusion algorithm is used to calculate the road congestion level index. By weighting and fusing distance and speed congestion information corrected for road type and curvature change rate, the influence of traffic flow, road geometric complexity, and road type is comprehensively considered, resulting in a more comprehensive and accurate index reflecting the current road congestion level. This provides a more precise input for subsequent assessments of driving scenario complexity. The weighted fusion algorithm can combine the two corrected factors according to preset weights to generate the final congestion level index.

[0126] Through the above technical solution, this application can more comprehensively assess road congestion levels by taking into account the impact of road geometry and road type on driving complexity, overcoming the limitations of assessing congestion solely based on vehicle spacing and speed. Therefore, the resulting congestion level index can more accurately reflect the complexity of actual driving scenarios, providing more reliable input for subsequent assessments of driving scenario complexity. This, in turn, supports the system in selecting more appropriate fault information presentation strategies in complex and ever-changing driving environments, improving the effectiveness of information presentation and driving safety.

[0127] In some preferred embodiments, the steps of calculating the curvature change rate for the road curvature sequence and correcting the distance congestion factor and the speed congestion factor according to the road type and the curvature change rate include:

[0128] Determine a reference curvature change rate threshold according to the road type of the road where the vehicle is located; the road types include highways, urban roads, and rural roads, and different road types correspond to different reference curvature change rate thresholds;

[0129] For the road curvature sequence, use a sliding window algorithm to calculate the curvature change rate sequence; the length of the sliding window is adaptively adjusted according to the vehicle speed, and the higher the vehicle speed, the longer the window length;

[0130] For the curvature change rate sequence, count the number of curvature change rates greater than the reference curvature change rate threshold to obtain the number of curvature change rates exceeding the threshold;

[0131] According to the road type and the number of curvature change rates exceeding the threshold, use a piecewise function to correct the distance congestion factor and the speed congestion factor; the correction amplitude for highways is less than that for urban roads and rural roads, and the more the number of curvature change rates exceeding the threshold, the greater the correction amplitude.

[0132] Among them, a reference curvature change rate threshold is set according to the road type determined by the current position of the vehicle. For example, a relatively high threshold can be set for highways, indicating that highways have a higher tolerance for curvature changes; a medium threshold can be set for urban roads; and a lower threshold can be set for rural roads, indicating that changes in the degree of curvature of rural roads are more likely to be perceived as an increase in driving difficulty. These thresholds can be preset based on actual driving data or expert experience and form a query table, and the reference curvature change rate threshold can be obtained by querying this query table according to the actual road type.

[0133] Furthermore, obtain the road curvature sequence within a period of time and use a sliding window algorithm to calculate the curvature change rate sequence. The length of the sliding window is not fixed but is adjusted according to the real-time speed of the vehicle. For example, when the vehicle speed is no greater than the first preset speed V1, the window length is set to L1; when the vehicle speed is greater than the second preset speed V2, the window length is L3; when the vehicle speed is between the first preset speed V1 and the second preset speed V2, the window length is L2; V1 < V2, L1 < L2 < L3, and each preset speed and window length can be set according to actual needs. This adaptive adjustment reflects that when driving at high speed, the driver needs a longer distance to perceive and respond to the road geometry changes ahead, so the curvature changes in a longer section need to be considered. The sliding window moves on the curvature sequence, calculates the change rate of the curvature within the window, and forms a curvature change rate sequence.

[0134] Therefore, for the calculated curvature change rate sequence, the number of points whose values ​​exceed the previously determined baseline curvature change rate threshold is counted. This number is the number of excess curvature change rates, which quantifies the "drastic" or "frequent" degree of road geometry change at the current speed. The more excess points, the more significant the curves or curvature changes in that road segment.

[0135] Finally, using road type and the number of curvature change rates exceeding the threshold obtained statistically, a piecewise function is used to correct the previously calculated distance congestion factor and speed congestion factor. The piecewise function can define different correction rules; for example, for highways, the correction magnitude is relatively small regardless of the number of exceeding the threshold; for urban and rural roads, the correction magnitude is larger, and the correction magnitude increases non-linearly (e.g., stepwise or piecewise linearly) as the number of exceeding the threshold curvature change rates increases. The specific piecewise function can be set according to actual needs and is not limited here. This correction method superimposes the additional influence of road geometry on driving difficulty and congestion perception onto the congestion factor calculated based on vehicle spacing and speed, making the congestion assessment results more reflective of the actual driving experience.

[0136] This correction method, which combines road type, speed adaptive window, and geometric complexity quantification, improves the accuracy of road congestion assessment and provides a more reliable basis for subsequent calculation of driving scenario complexity.

[0137] Preferably, step A203 may include:

[0138] Based on driver operation data, throttle opening sequence, brake pressure sequence and steering torque sequence are extracted, and the mean and variance of each sequence are calculated.

[0139] Based on driver attention-related data, blink frequency from eye movement data and head-turning frequency from head posture data were extracted.

[0140] For the mean and variance of the throttle opening sequence, the mean and variance of the braking pressure sequence, the mean and variance of the steering torque sequence, the blink frequency and the nodding frequency, a preset fuzzy inference rule is used to calculate the driver fatigue level index and the driver distraction level index.

[0141] This method provides a specific implementation for assessing driver fatigue and distraction levels.

[0142] First, based on driver operation data, throttle opening sequence, brake pressure sequence, and steering torque sequence are extracted. The throttle opening sequence reflects the driver's control input regarding the vehicle's acceleration or deceleration intention. The brake pressure sequence reflects the driver's control input regarding the vehicle's deceleration or stopping intention. The steering torque sequence reflects the driver's input regarding vehicle directional control. The mean and variance are calculated for these sequences. The mean reflects the average level or trend of the operation, such as average throttle opening or average steering torque. The variance reflects the volatility or stationarity of the operation; for example, the more unstable the operation, the larger the variance. Thus, the raw operation data is transformed into quantifiable statistical features that characterize the driver's behavioral patterns when controlling the vehicle.

[0143] Furthermore, based on driver attention-related data, blink frequency from eye movement data and head-shaking frequency from head posture data were extracted. Eye movement and head posture data were obtained through image recognition. Blink frequency is the number of times the driver blinks per unit time; abnormal blinking patterns are associated with fatigue. Head posture data reflects the driver's head orientation and movement; head-shaking frequency is the number of times the head shakes per unit time; frequent or abnormal head movements are associated with distraction. Extracting these specific frequency indicators allows for the direct acquisition of behavioral characteristics related to the driver's physiological state and attention.

[0144] Specifically, the extracted feature values ​​are used as input to a fuzzy inference system. This system contains a pre-defined fuzzy rule base, which, based on expert knowledge or training data, describes the relationship between input features and fatigue and distraction levels. For example, a rule might state that "high blinking frequency and large steering torque variance indicate high fatigue." Through fuzzification, fuzzy inference, and defuzzification processes, these multi-source, potentially uncertain, input features are synthesized to output quantified indicators of driver fatigue and distraction levels. This method can handle the complex relationships and uncertainties between input features, providing a comprehensive evaluation result.

[0145] Through the above technical solution, this application solves the problems of unclear driver fatigue and distraction assessment methods and potentially inaccurate or unstable assessment results in existing technologies. By extracting the mean and variance of throttle opening, braking pressure, and steering torque, as well as blink frequency and head sway frequency—multiple features closely related to the driver's state—from driver operation data and attention-related data, and using these multi-source heterogeneous features as input, fuzzy inference rules are employed to comprehensively analyze and judge these features. This simulates the fuzzy judgment process of humans in complex situations, handling uncertainties and nonlinear relationships in the data. This method based on multi-feature fusion and fuzzy inference can more comprehensively and robustly assess the driver's fatigue and distraction levels, obtaining more reliable fatigue and distraction indicators. Improved accuracy of the assessment results directly enhances the reliability of subsequent driver cognitive load calculations, thereby supporting the system in more accurately selecting information presentation strategies. This helps avoid presenting too much information when the driver is in a poor state, increasing their burden, or presenting insufficient information when the driver is in a good state, affecting comprehension. This improves the adaptability and effectiveness of fault information presentation, enhancing the safety of human-computer interaction and user experience.

[0146] Preferably, step A204 may include:

[0147] The stability index, congestion index, fatigue index, and distraction index were normalized.

[0148] The driving scenario complexity is obtained by weighted fusion calculation of the difference between 1 and the normalized stability index and the normalized congestion index.

[0149] The driver's cognitive load is obtained by weighted fusion calculation of the normalized fatigue level index and the normalized distraction level index.

[0150] This solution proposes a specific calculation method to address the problem of how to comprehensively consider stability indicators, congestion indicators, fatigue indicators, and distraction indicators to calculate the complexity of driving scenarios and the cognitive load of drivers.

[0151] First, the stability index, congestion index, fatigue index, and distraction index are normalized. Normalization transforms the values ​​of different indicators to a uniform scale, for example, mapping all index values ​​to the range of 0 to 1. As a preferred implementation method, the min-maximum normalization method can be used, subtracting the historical minimum value from the current value of each indicator, and then dividing by the difference between the historical maximum and minimum values ​​to obtain a normalized value between 0 and 1. This eliminates the differences in numerical range or units among different indicators, making subsequent comprehensive calculations more reasonable.

[0152] Next, when calculating the complexity of the driving scenario, a weighted fusion calculation is performed using the difference between 1 and the normalized stability index, and the normalized congestion index. The stability index reflects the stability of vehicle driving; a higher stability index value indicates a smoother driving experience. The difference between 1 and the normalized stability index represents lower stability; a larger difference reflects greater vehicle instability and higher driving scenario complexity. The normalized congestion index directly reflects the degree of road congestion; higher congestion levels indicate higher driving scenario complexity. By weighting and fusing these two factors, for example using a linear weighted summation, the influence of the vehicle's own driving state (stability) and the external traffic environment (congestion level) on the complexity of the driving scenario can be comprehensively considered. For example, the complexity of the driving scenario can be calculated as W1*(1-normalized stability index) + W2*normalized congestion index, where W1 and W2 are weighting coefficients that can be calibrated based on actual driving data or determined through expert experience. Weighted fusion allows assigning different weights to the two components based on the actual situation, making the calculation results more realistic.

[0153] Finally, when calculating driver cognitive load, a weighted fusion method is used to calculate the normalized fatigue level index and the normalized distraction level index. The normalized fatigue level index reflects the driver's physiological fatigue state; the higher the fatigue level, the higher the cognitive load. The normalized distraction level index reflects the driver's degree of attentional inattention; the higher the distraction level, the higher the cognitive load. By weighting and fusing these two indicators, for example using a linear weighted summation, the impact of the driver's current physiological and psychological state on their cognitive abilities can be comprehensively assessed. For example, driver cognitive load can be calculated as W3 * normalized fatigue level index + W4 * normalized distraction level index, where W3 and W4 are weighting coefficients that can be calibrated based on actual driving data or determined through expert experience. The weighted fusion also allows assigning different weights to the two indicators based on actual circumstances, making the calculation results more accurately reflect the driver's current cognitive load level.

[0154] Through the above technical solutions, this application eliminates the differences in dimensions and numerical ranges among different indicators, making the stability index, congestion index, fatigue index, and distraction index comparable. By weightedly fusing the difference between 1 and the normalized stability index with the normalized congestion index, the influence of vehicle driving status and external traffic environment on the complexity of the driving scenario can be comprehensively considered, obtaining an assessment result reflecting the complexity of the driving scenario. By weightedly fusing the normalized fatigue index and the normalized distraction index, the influence of the driver's physiological and psychological state on cognitive load can be comprehensively reflected, obtaining an assessment result reflecting the driver's cognitive load. Therefore, more accurate and stable assessment results of driving scenario complexity and driver cognitive load can be obtained than directly using the original indicators, providing a reliable basis for the selection of subsequent information presentation strategies.

[0155] In some implementations, step A4 includes:

[0156] A401. Prioritize fault information based on fault type, fault severity, and vehicle performance impact assessment results;

[0157] A402. Assess the risk level of a driving scenario based on its complexity;

[0158] A403. Determine the urgency of information presentation based on the priority of fault information and the risk level of the driving scenario;

[0159] A404. Select an information presentation strategy from a preset strategy library based on the urgency of the information presentation and the driver's cognitive load; the information presentation strategy includes the information presentation channel, the level of detail of the information content, and the timing of presentation.

[0160] Step A401 processes the received fault information, converting the inherent attributes of the fault and its impact on vehicle functions into a priority value. For example, a fault affecting braking function is usually given a higher priority than a fault affecting air conditioning function. Fault types can include motor faults, battery faults, braking system faults, etc. Fault severity can be categorized as minor, moderate, and severe. The vehicle performance impact assessment results can quantify the impact of the fault on power output, driving range, braking performance, etc. By combining these factors, a priority index reflecting the importance of the fault can be obtained. As a preferred implementation, a lookup table method or rule-based reasoning method can be used to determine the priority. For example, a rule set can be defined: if the fault type is a braking system fault and the severity is severe, the priority is the highest; if the fault type is a battery fault and the vehicle performance impact assessment results show a significant decrease in driving range, the priority is high.

[0161] Furthermore, in step A402, based on the assessment of the complexity of the driving scenario, this step correlates the complexity with potential driving risks. For example, the complexity of a scenario driving in congested urban areas may be high, but due to the lower vehicle speed, its risk level may be lower than that of a scenario driving at high speeds on a highway, even if the complexity (e.g., traffic flow) is similar on the highway. The risk level reflects the potential challenges to the driver's safe operation in the current driving environment. The assessment can be based on a pre-defined risk model that maps different levels of driving scenario complexity to different risk levels; for example, classifying complexity into low, medium, and high, and corresponding to low, medium, and high risk levels. Alternatively, the risk level can be dynamically calculated based on specific environmental perception data (e.g., surrounding vehicle density, relative speed, road curvature change rate) and vehicle operation data (e.g., vehicle speed). For example, when the vehicle speed exceeds a certain threshold and the surrounding vehicle density is high, the risk level is assessed as high.

[0162] Therefore, step A403 comprehensively considers the urgency of the fault itself and the danger of the current driving environment. A high-priority fault may need to be presented immediately in a low-risk scenario, while in a high-risk scenario it may need to be delayed or presented in a more cautious manner. Conversely, a low-priority fault may not need to be presented immediately even in a low-risk scenario. The urgency of information presentation reflects the urgency with which the system conveys fault information to the driver. This urgency can be determined using a two-dimensional lookup table or a decision matrix, where one dimension is the fault information priority and the other is the driving scenario risk level. The cells of the table or matrix store the corresponding information presentation urgency values. For example, when the fault information priority is high and the driving scenario risk level is low, the information presentation urgency is determined to be the highest; when the fault information priority is high and the driving scenario risk level is high, the information presentation urgency may be determined to be medium or low, or it may need to be delayed. In this way, the system can avoid distracting the driver's attention with information presentation during high-risk moments.

[0163] Specifically, step A404 selects an information presentation strategy from a preset strategy library based on the urgency of the information presentation and the driver's cognitive load. After determining the overall urgency of the information presentation, and considering the driver's current cognitive load (which has been assessed in previous steps), the system selects the most suitable strategy from a preset library containing multiple information presentation strategies. Strategies in the library can be indexed based on different combinations of information presentation urgency and driver cognitive load. For example, when the information presentation urgency is highest and the driver's cognitive load is low, detailed fault information can be presented simultaneously through multiple channels—visual, auditory, and tactile—and immediately. When the information presentation urgency is moderate and the driver's cognitive load is high, a brief warning message can be presented only through the visual channel, with a delayed presentation until the cognitive load decreases. Information presentation channels can include at least one of the following: instrument panel display, central control screen display, voice prompts, warning sounds, seat vibrations, etc. The level of detail in the information content can range from simple warning lights to detailed fault descriptions and suggested actions. The presentation timing can be immediate, delayed for a certain period, or under specific conditions (such as reduced vehicle speed or reduced driver cognitive load). By comprehensively considering the urgency of the situation and cognitive load, the system can select a strategy that ensures effective information delivery while minimizing interference with the driver's safe operation. This step, combined with the assessment of the complexity of the driving scenario and the driver's cognitive load, allows the selection of information presentation strategies to adapt to dynamically changing driving environments and driver states. This enables a more intelligent balance between information delivery needs and driving safety requirements in the event of a malfunction.

[0164] Through the above technical solution, this application refines the evaluation dimensions of fault information and introduces fault information priority; by transforming the complexity of the driving scenario into a driving scenario risk level, it more directly reflects the impact of the environment on driving safety; by comprehensively considering the fault information priority and the driving scenario risk level, it determines the overall urgency of information presentation, thereby establishing a correlation between the urgency of the fault itself and the risk level of the current driving environment; finally, by combining the driver's cognitive load, it selects the most suitable information presentation strategy from a preset strategy library. Therefore, this application can select information presentation strategies more precisely and intelligently, effectively balancing the timeliness of fault information delivery with the need to avoid interfering with the driver's safe operation, solving the problem of insufficiently optimized strategy selection in existing solutions, improving the adaptability and effectiveness of fault information presentation, and helping to ensure driving safety and improve user experience.

[0165] refer to Figure 2 This application provides a fault handling system for collaborative control of new energy vehicles, used to present fault information to the driver when a fault occurs in the collaborative control system of a new energy vehicle during vehicle operation. The system includes:

[0166] Data acquisition module 1 is used to acquire vehicle operation data, environmental perception data, driver operation data and driver attention-related data (for details, refer to step A1 above).

[0167] Evaluation module 2 is used to evaluate the complexity of the driving scenario and the cognitive load of the driver based on the acquired vehicle operation data, environmental perception data, driver operation data and driver attention-related data (refer to step A2 above for the specific process).

[0168] The fault information receiving module 3 is used to receive fault information output by the cooperative control system; the fault information includes fault type, fault severity and vehicle performance impact assessment results (refer to step A3 above for details).

[0169] The strategy selection module 4 is used to select an information presentation strategy from a preset strategy library based on fault information, the complexity of the driving scenario, and the driver's cognitive load. The information presentation strategy includes the information presentation channel, the level of detail of the information content, and the timing of presentation (refer to step A4 above for details).

[0170] Information presentation module 5 is used to generate fault information according to the selected information presentation strategy and control the human-machine interface device to perform information presentation (for details, refer to step A5 above).

[0171] The above description is merely an embodiment of this application and is not intended to limit the scope of protection of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for handling faults in the collaborative control system of a new energy vehicle, used to present fault information to the driver when a fault occurs in the collaborative control system of a new energy vehicle during vehicle operation, characterized in that, The steps of this method include: A1. Acquire vehicle operation data, environmental perception data, driver operation data, and driver attention-related data; A2. Based on the acquired vehicle operation data, environmental perception data, driver operation data, and driver attention-related data, assess the complexity of the driving scenario and the driver's cognitive load; A3. Receive fault information output by the cooperative control system; the fault information includes fault type, fault severity, and vehicle performance impact assessment results; A4. Based on fault information, the complexity of the driving scenario, and the driver's cognitive load, select an information presentation strategy from the preset strategy library; the information presentation strategy includes the information presentation channel, the level of detail of the information content, and the timing of presentation. A5. Based on the selected information presentation strategy, generate fault information and control the human-machine interface device to perform information presentation; Step A1 includes: A101. Read vehicle operation data via the vehicle bus CAN interface; vehicle operation data includes vehicle speed, steering angle, and acceleration; A102. Acquire environmental perception data through radar and cameras; environmental perception data includes distance to surrounding vehicles, relative speed, and road curvature. A103. Acquire driver operation data through sensors; driver operation data includes throttle opening, brake pressure, and steering torque; A104. Obtain driver attention-related data through image recognition; driver attention-related data includes eye movement data and head posture. A105. Preprocess vehicle operation data, environmental perception data, driver operation data, and driver attention-related data; Step A2 includes: A201. Based on vehicle operation data, assess the vehicle's driving stability and obtain a stability index; A202. Based on environmental perception data and road type, assess the degree of road congestion to obtain congestion level indicators; A203. Based on driver operation data and driver attention-related data, assess the driver's fatigue level and distraction level to obtain fatigue level index and distraction level index; A204. Calculate the complexity of driving scenarios and the cognitive load of drivers by combining stability index, congestion index, fatigue index, and distraction index. Step A202 includes: The road type of the road where the vehicle is located is determined based on the vehicle's real-time location; Based on environmental perception data, extract the distance sequence of surrounding vehicles, the relative speed sequence, and the road curvature sequence; For the distance sequence and relative speed sequence of surrounding vehicles, a dynamic time warping algorithm is used to calculate the distance congestion factor and speed congestion factor; For the road curvature sequence, calculate the rate of change of curvature, and adjust the distance congestion factor and speed congestion factor according to the road type and the rate of change of curvature. Based on the corrected distance congestion factor and the corrected speed congestion factor, a weighted fusion algorithm is used to calculate the road congestion index.

2. The method for handling faults in the collaborative control of new energy vehicles according to claim 1, characterized in that, Step A201 includes: Based on vehicle operation data, extract vehicle speed sequence, steering angle sequence, and acceleration sequence; For the vehicle speed sequence, steering angle sequence, and acceleration sequence, calculate the sequence entropy to obtain the vehicle speed sequence entropy, steering angle sequence entropy, and acceleration sequence entropy. Based on the vehicle speed sequence entropy, steering angle sequence entropy, and acceleration sequence entropy, a weighted fusion algorithm is used to calculate the stability index.

3. The method for handling faults in the collaborative control of new energy vehicles according to claim 1, characterized in that, The steps of calculating the rate of change of curvature for a road curvature sequence and correcting the distance congestion factor and speed congestion factor based on the road type and the rate of change of curvature include: The reference curvature change rate threshold is determined based on the road type where the vehicle is located; road types include highways, urban roads and rural roads, and different road types correspond to different reference curvature change rate thresholds. For road curvature sequences, a sliding window algorithm is used to calculate the curvature change rate sequence; the length of the sliding window is adaptively adjusted according to the vehicle speed, and the higher the vehicle speed, the longer the window length. For the curvature change rate sequence, count the number of curvature change rates that are greater than the baseline curvature change rate threshold to obtain the number of curvature change rates exceeding the threshold. Based on the road type and the number of curvature change rates exceeding the threshold, a piecewise function is used to correct the distance congestion factor and the speed congestion factor; the correction range for highways is smaller than that for urban roads and rural roads, and the more curvature change rates exceeding the threshold, the greater the correction range.

4. The method for handling faults in the collaborative control of new energy vehicles according to claim 1, characterized in that, Step A203 includes: Based on driver operation data, throttle opening sequence, brake pressure sequence and steering torque sequence are extracted, and the mean and variance of each sequence are calculated. Based on driver attention-related data, blink frequency from eye movement data and head-turning frequency from head posture data were extracted. For the mean and variance of the throttle opening sequence, the mean and variance of the braking pressure sequence, the mean and variance of the steering torque sequence, the blink frequency and the nodding frequency, a preset fuzzy inference rule is used to calculate the driver fatigue level index and the driver distraction level index.

5. The method for handling faults in the collaborative control of new energy vehicles according to claim 1, characterized in that, Step A204 includes: The stability index, congestion index, fatigue index, and distraction index were normalized. The driving scenario complexity is obtained by weighted fusion calculation of the difference between 1 and the normalized stability index and the normalized congestion index. The driver's cognitive load is obtained by weighted fusion calculation of the normalized fatigue level index and the normalized distraction level index.

6. The method for handling faults in the collaborative control of new energy vehicles according to claim 1, characterized in that, Step A4 includes: A401. Prioritize fault information based on fault type, fault severity, and vehicle performance impact assessment results; A402. Assess the risk level of a driving scenario based on its complexity; A403. Determine the urgency of information presentation based on the priority of fault information and the risk level of the driving scenario; A404. Select an information presentation strategy from a preset strategy library based on the urgency of the information presentation and the driver's cognitive load; the information presentation strategy includes the information presentation channel, the level of detail of the information content, and the timing of presentation.

7. A fault handling system for collaborative control of new energy vehicles, used to present fault information to the driver when a fault occurs in the collaborative control system of a new energy vehicle during vehicle operation, characterized in that, The system includes: The data acquisition module is used to acquire vehicle operation data, environmental perception data, driver operation data, and driver attention-related data. The evaluation module is used to assess the complexity of the driving scenario and the cognitive load of the driver based on the acquired vehicle operation data, environmental perception data, driver operation data, and driver attention-related data. The fault information receiving module is used to receive fault information output by the cooperative control system; the fault information includes fault type, fault severity and vehicle performance impact assessment results; The strategy selection module is used to select an information presentation strategy from a preset strategy library based on fault information, the complexity of the driving scenario, and the driver's cognitive load. The information presentation strategy includes the information presentation channel, the level of detail of the information content, and the timing of presentation. The information presentation module is used to generate fault information based on the selected information presentation strategy and control the human-machine interface device to perform information presentation. The data acquisition module is also used for: Vehicle operating data is read via the vehicle bus CAN interface; vehicle operating data includes vehicle speed, steering angle, and acceleration. Environmental perception data is acquired through radar and cameras; this data includes distances to surrounding vehicles, relative speeds, and road curvature. Driver operation data is acquired through sensors; driver operation data includes throttle opening, brake pressure, and steering torque. Driver attention-related data is obtained through image recognition; this data includes eye movement data and head posture. Preprocess vehicle operation data, environmental perception data, driver operation data, and driver attention-related data; The evaluation module is also used for: Based on vehicle operation data, the vehicle's driving stability is assessed to obtain a stability index; Based on environmental perception data and road type, assess the degree of road congestion to obtain a congestion level index; Based on driver operation data and driver attention-related data, the driver's fatigue level and distraction level are assessed, and fatigue level index and distraction level index are obtained. By combining stability indicators, congestion indicators, fatigue indicators, and distraction indicators, the complexity of driving scenarios and the cognitive load of drivers are calculated. Based on environmental perception data and road type, the degree of road congestion is assessed, and the congestion level indicators include: The road type of the road where the vehicle is located is determined based on the vehicle's real-time location; Based on environmental perception data, extract the distance sequence of surrounding vehicles, the relative speed sequence, and the road curvature sequence; For the distance sequence and relative speed sequence of surrounding vehicles, a dynamic time warping algorithm is used to calculate the distance congestion factor and speed congestion factor; For the road curvature sequence, calculate the rate of change of curvature, and adjust the distance congestion factor and speed congestion factor according to the road type and the rate of change of curvature. Based on the corrected distance congestion factor and the corrected speed congestion factor, a weighted fusion algorithm is used to calculate the road congestion index.

Images