Vehicle control method and device and vehicle

Abstract

The invention relates to the field of vehicle intelligent control, in particular to a vehicle control method and device and a vehicle. Compared with a traditional mode depending on visual perception and a machine learning model, the driving state of the driver and the vehicle driving risk data are detected at the same time, the misjudgment limitation of single perception is avoided, for example, potential risks can be judged in an auxiliary mode through abnormal states such as distraction and fatigue of the driver. Secondly, whether active safety intervention is triggered or not can be determined by analyzing driving risk data in real time, the passive situation that a traditional algorithm focuses on protection after collision is changed, and real-time monitoring of danger in the driving process is achieved. The target intervention strategy is matched according to the driving state and the driving risk data, corresponding safety operation is executed, a complete closed loop from detection and analysis to active intervention is formed, the risk is actively reduced before an accident occurs, and the defect that the traditional technology lacks an active prevention mechanism is overcome.

Description

Technical Field

[0001] The present invention relates to the field of vehicle intelligent control, and in particular to a vehicle control method, device and vehicle. Background Art

[0002] Existing vehicle autonomous driving systems primarily rely on visual perception and machine learning models. These systems, limited by factors such as environmental interference and data bias, carry a high risk of misjudgment, making it difficult to accurately identify potential hazards in complex scenarios. Furthermore, traditional automotive safety control algorithms mostly focus on passive protection after a collision (such as airbags and energy absorption by the vehicle body structure). They lack real-time monitoring and prevention mechanisms for hazards that arise during driving, making it impossible to proactively mitigate risks or avoid collisions before an accident occurs. Summary of the Invention

[0003] In view of this, embodiments of the present invention provide a vehicle control method, device, and vehicle to address the problems that existing autonomous driving systems have the risk of misjudgment due to their reliance on visual perception and machine learning models, and that traditional safety control algorithms lack real-time monitoring and active prevention mechanisms for driving hazards.

[0004] In a first aspect, an embodiment of the present invention provides a vehicle control method, the method comprising:

[0005] During vehicle driving, detect the driver's driving status and obtain vehicle driving risk data;

[0006] Analyzing the driving risk data to determine whether to trigger active safety intervention for the vehicle during driving;

[0007] If active safety intervention is triggered for the vehicle, obtaining a target intervention strategy matched with the driving state and / or the driving risk data;

[0008] Executing a safety intervention operation corresponding to the target intervention strategy on the vehicle.

[0009] Furthermore, the obtaining of the vehicle's driving risk data includes:

[0010] Detecting a target object that is in conflict with the vehicle, and obtaining a remaining collision time between the vehicle and the target object;

[0011] Obtaining the shortest braking distance required for the vehicle to be completely braked from the current speed to a standstill;

[0012] Obtaining the number of interactions between the driver and the vehicle's interactive system within a unit time;

[0013] The remaining time to collision, the shortest braking distance, and the number of interactions are used as the driving risk data.

[0014] Furthermore, analyzing the driving risk data to determine whether to trigger active safety intervention for the vehicle during driving includes:

[0015] Performing a weighted sum based on the remaining collision time, the shortest braking distance, and the number of interactions to obtain risk quantification data;

[0016] Comparing the risk quantification data with a first threshold to obtain a first comparison result;

[0017] Determine whether to trigger active safety intervention for the vehicle during driving based on the first comparison result, wherein, if the first comparison result is that the risk quantification data is less than the first threshold, active safety intervention is not triggered for the vehicle during driving; or, if the first comparison result is that the risk quantification data is greater than or equal to the first threshold, active safety intervention is triggered for the vehicle during driving.

[0018] Furthermore, the target intervention strategy for obtaining the driving status and / or the driving risk data includes:

[0019] If the driving state is an unsafe driving state, and / or the risk quantification data is less than a second threshold and greater than or equal to a first threshold, determining that the targeted intervention strategy is the first intervention strategy, wherein the risk quantification data is obtained by weighted summation of the remaining time to collision, the shortest braking distance, and the number of interactions; or

[0020] If the remaining time to collision in the driving risk data is less than the preset time and the driving state is an unsafe driving state, determining that the target intervention strategy to be hit is the second intervention strategy; or,

[0021] If the risk quantification data is greater than or equal to the second threshold, and / or the first intervention strategy or the second intervention strategy does not take effect after execution, it is determined that the hit target intervention strategy is the third intervention strategy.

[0022] Furthermore, executing the safety intervention operation corresponding to the target intervention strategy on the vehicle includes:

[0023] If the target intervention strategy is the first intervention strategy, intervention signals are sent to the sound and light alarm device, seat vibration device and display device located inside the vehicle respectively, so as to control the sound and light alarm device to perform sound and light alarm operations, control the seat vibration device to perform vibration operations and control the display device to display warning prompt information through the intervention signal.

[0024] Furthermore, executing the safety intervention operation corresponding to the target intervention strategy on the vehicle includes:

[0025] If the target intervention strategy is the second intervention strategy, the power output of the vehicle is limited, and the rotation angle of the vehicle steering wheel is limited to a set angle range.

[0026] Furthermore, executing the safety intervention operation corresponding to the target intervention strategy on the vehicle includes:

[0027] If the target intervention strategy is the third intervention strategy, controlling the vehicle to perform emergency braking, and obtaining the number and type of target objects that have a driving conflict with the vehicle;

[0028] Planning a safe driving path for the vehicle based on the number of objects and the types of objects;

[0029] The vehicle is controlled to travel along the safe driving path.

[0030] Furthermore, planning a safe driving path for the vehicle based on the number of objects and the types of objects includes:

[0031] planning at least one candidate driving path based on the number of objects;

[0032] Obtaining an object weight corresponding to the object type, wherein the object weight is used to indicate a degree of protection of different types of objects when a driving conflict occurs with the vehicle;

[0033] Calculating a risk value for each candidate driving path using the number of objects and the object weights;

[0034] The candidate driving path with the smallest risk value is used as the safe driving path for the vehicle.

[0035] In a second aspect, an embodiment of the present invention provides a vehicle control device, the device comprising:

[0036] The detection module is used to detect the driver's driving status and obtain the vehicle's driving risk data during the vehicle's driving process;

[0037] An analysis module, configured to analyze the driving risk data and determine whether to trigger active safety intervention for the vehicle during driving;

[0038] an acquisition module, configured to acquire a target intervention strategy hit by the driving state and / or the driving risk data if active safety intervention is triggered for the vehicle;

[0039] An execution module is used to execute the safety intervention operation corresponding to the target intervention strategy on the vehicle.

[0040] In a third aspect, an embodiment of the present invention provides a vehicle, comprising: a memory and a controller, wherein the memory and the controller are communicatively connected to each other, computer instructions are stored in the memory, and the controller executes the above method by executing the computer instructions.

[0041] In a fourth aspect, an embodiment of the present invention provides an electronic device comprising: a memory and a processor, the memory and the processor being communicatively connected to each other, computer instructions being stored in the memory, and the processor executing the method of the first aspect or any corresponding embodiment thereof by executing the computer instructions.

[0042] In a fifth aspect, an embodiment of the present invention provides a computer-readable storage medium having computer instructions stored thereon, the computer instructions being used to enable a computer to execute the method of the first aspect or any corresponding embodiment thereof.

[0043] The embodiments of the present application effectively overcome the deficiencies of the existing technology through multi-dimensional data collection and dynamic response mechanisms. Compared with the traditional method that relies on visual perception and machine learning models, the present application simultaneously detects the driver's driving status and vehicle driving risk data to avoid the misjudgment limitations of a single perception. For example, it can assist in judging potential risks through abnormal conditions such as driver distraction and fatigue. Secondly, by analyzing driving risk data in real time, it can determine whether to trigger active safety intervention, change the passive situation of traditional algorithms that focus on post-collision protection, and realize real-time monitoring of driving dangers. According to the driving status and driving risk data, the target intervention strategy is matched, and corresponding safety operations are performed to form a complete closed loop from detection, analysis to active intervention, actively reducing risks before an accident occurs, and making up for the lack of active prevention mechanisms in traditional technologies. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the specific embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0045] Figure 1 is a flow chart of a vehicle control method according to some embodiments of the present invention;

[0046] Figure 2 is a flow chart of another vehicle control method according to some embodiments of the present invention;

[0047] Figure 3is a flow chart of another vehicle control method according to some embodiments of the present invention;

[0048] Figure 4 is a flow chart of another vehicle control method according to some embodiments of the present invention;

[0049] Figure 5 is a structural block diagram of a vehicle control device according to an embodiment of the present invention;

[0050] Figure 6 FIG. 4 is a schematic diagram of the hardware structure of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0051] To make the purpose, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making creative efforts shall fall within the scope of protection of the present invention.

[0052] According to an embodiment of the present invention, a vehicle control method, device and vehicle are provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0053] In this embodiment, a vehicle control method is provided. Figure 1 is a flow chart of a vehicle control method according to an embodiment of the present invention. Figure 1 As shown, the process includes the following steps:

[0054] Step S101 : During vehicle driving, the driver's driving state is detected and driving risk data of the vehicle is obtained.

[0055] In an embodiment of the present application, a multi-dimensional sensor array is used to collect the driver's physiological and behavioral data in real time: a seat pressure sensor array (16-point matrix) is distributed in the seat basin, backrest and armrest areas, and a piezoresistive sensor is used to monitor the body pressure distribution to determine whether the driver maintains a normal driving posture (such as abnormal conditions such as sitting posture deviation and leaving the seat); the steering wheel bioelectric sensor is integrated on the inside of the skeleton (avoiding the airbag path), and an ECG electrode array is used to collect biological signals such as heart rate and micro-expressions to identify physiological fatigue or tension; the binocular eye tracking camera above the center console uses stereo vision technology to monitor eye focus (such as whether it deviates from the road), blinking frequency and other eye movement characteristics to determine whether there is distraction (such as the line of sight deviating from the road for more than 3 seconds) or signs of fatigue.

[0056] All sensor data is transmitted to the domain controller via the CAN bus. After filtering and denoising, the LSTM machine learning model is used for feature extraction to identify unsafe driving conditions such as distraction, fatigue, and abnormal expressions.

[0057] Specifically, a multimodal dataset is first constructed, collecting driver data (such as eye movement trajectories, heart rate variability, and seat pressure distribution) under normal driving conditions as baseline samples. Feature samples indicating distraction (such as looking at a phone or talking with the head tilted), fatigue (such as frequent blinking and lowered head), and abnormal facial expressions (such as anger and panic) are also annotated. The raw sensor data is preprocessed: bioelectric signals are decomposed using wavelet transforms to extract heart rate variability features, facial micro-expression features are extracted using the HOG (Histogram of Oriented Gradients) algorithm, and seat pressure data is converted into a spatiotemporal heat map sequence. Training samples are constructed using a sliding time window (e.g., 3 seconds in length and 50% overlap), each containing multi-channel time series data (eye movement angle, blink frequency, heart rate value, pressure distribution matrix, etc.). The preprocessed samples were fed into an LSTM model. The model structure consists of three layers of bidirectional LSTM units (128 neurons per layer) to capture long- and short-term temporal dependencies. An attention mechanism automatically weights the importance of features at different time steps. Finally, a fully connected layer is connected to a Softmax classifier to output probability distributions for states such as distracted, tired, and normal. The model was trained using the Adam optimizer and the cross-entropy loss function, with early stopping and dropout (0.3) used to prevent overfitting. Ultimately, the model achieved state recognition accuracy exceeding 92% on the validation set.

[0058] The domain controller receives preprocessed sensor data streams in real time and feeds a continuous 3-second data window (containing 600 time steps, sampled at 200Hz) into a trained LSTM model. The model first extracts temporal features through a bidirectional LSTM layer. For example, it detects features such as a gaze deviation of >30° from the road center for more than 2 seconds, a heart rate less than 50 beats / minute, and a blink rate >20 times / minute. The attention mechanism automatically focuses on feature vectors at key time points (such as when an unusual expression appears). The softmax layer outputs a probability distribution of the current driving state. When the probability of a distracted state is >0.8 and the duration is >1 second, the system determines that the driver is distracted.

[0059] To avoid misjudgments, a state confirmation mechanism is implemented. This means that a warning is triggered only when the same unsafe state is detected at least twice in three consecutive time windows (a total of 4.5 seconds). Simultaneously, the model dynamically updates the driving behavior baseline, adjusting fatigue thresholds based on the driver's personal habits (for example, a driver who frequently stays up late may have an elevated heart rate baseline). Ultimately, the domain controller sends the recognition result (e.g., "distracted state, 95% confidence") to the decision-making module, triggering the appropriate level of safety intervention.

[0060] In the embodiment of the present application, obtaining the driving risk data of the vehicle includes the following steps A1-A4:

[0061] Step A1: Detect a target object that is in conflict with the vehicle and obtain a remaining collision time between the vehicle and the target object.

[0062] Specifically, the sensor fusion device on the front of the vehicle (integrated with a binocular camera, lidar, and millimeter-wave radar) scans the environment in front of the vehicle in real time. The millimeter-wave radar detects the distance between the vehicle and the target object with an accuracy of ±10cm. The lidar and camera jointly identify the target object (such as a pedestrian or vehicle) and track its movement trajectory, calculating the relative speed of the vehicle and the target object. Based on the formula TTC = distance / relative speed, the remaining time to collision (TTC) is output in real time.

[0063] When it is detected that the target object's motion trajectory and the vehicle's driving path have an intersecting or approaching trend (i.e., driving conflict), TTC calculation is triggered and the data is continuously updated.

[0064] Step A2: Obtain the shortest braking distance required for the vehicle to be completely braked from the current speed to a standstill.

[0065] Specifically, according to the vehicle's current speed v, the road friction coefficient μ (monitored in real time by on-board sensors or preset based on weather data, such as μ≈0.8 on dry roads and μ≈0.3 on wet roads), and the braking performance parameters (comprehensive coefficients such as brake pad friction and air resistance obtained through actual vehicle calibration), the formula DDT=v is used. 2 / (2×g×μ)(g is the acceleration due to gravity, take 9.8m / s 2 )Calculate the shortest braking distance.

[0066] The vehicle speed signal transmitted by the CAN bus is acquired in real time, and combined with the road condition data output by the environmental perception module (such as slippery road parameters triggered by rainy and snowy weather), the DDT calculation results are dynamically adjusted to reflect the minimum space redundancy required for the vehicle to detect danger and come to a complete stop in its current state.

[0067] Step A3: Obtain the number of interactions between the driver and the vehicle's interactive system within a unit time.

[0068] Specifically, the operation log recording module of the in-vehicle HMI interactive system (such as the instrument panel screen, steering wheel buttons, voice control, etc.) is used to count the number of interactions between the driver and the system per unit time (such as per second, per minute) in real time.

[0069] Interaction types include but are not limited to touch screen operations, pressing steering wheel buttons, voice command input, etc. The system obtains the timestamp of the interaction event through the CAN bus or in-vehicle Ethernet, accumulates the number of interactions according to a preset time window (such as 1 minute), and forms real-time interaction frequency data between the driver and the vehicle interaction system (such as HMI interaction frequency = number of interactions / time), which is used to assess the impact of driver distraction on driving safety.

[0070] In step A4, the remaining time to collision, the shortest braking distance, and the number of interactions are used as driving risk data.

[0071] Specifically, by detecting objects in conflict with the vehicle and obtaining the remaining time to collision, the system can predict the time urgency of a potential collision. The system also obtains the minimum braking distance required for a full vehicle braking, assessing the feasibility of avoiding a collision from a spatial perspective. Combined with the remaining time to collision, this system forms a dual-dimensional risk assessment system in both time and space. The system then quantifies the driver's attention to the vehicle, combining the number of driver interactions to determine whether their driving behavior is safe. The remaining time to collision, minimum braking distance, and number of interactions are combined into driving risk data, enabling a fusion analysis of multi-source information during vehicle operation.

[0072] Step S102: analyzing the driving risk data to determine whether active safety intervention is triggered for the vehicle during driving.

[0073] In the embodiment of the present application, analyzing the driving risk data and determining whether to trigger active safety intervention for the vehicle during driving includes the following steps B1-B3:

[0074] Step B1: performing weighted summation based on the remaining collision time, the shortest braking distance, and the number of interactions to obtain risk quantification data.

[0075] Specifically, the obtained remaining time to collision (TTC), shortest braking distance (DDT) and number of interactions per unit time are weighted and summed according to preset weights.

[0076] The formula is: Risk Quantification Data (Risk Quantification Data) = 0.6 × TTC + 0.3 × DDT + 0.1 × Number of Interactions. TTC reflects the time urgency of a collision (weight 0.6), DDT reflects the spatial margin required for vehicle braking (weight 0.3), and the number of interactions represents the degree of driver distraction (weight 0.1). Based on this, the multi-dimensional risk parameters are integrated into a unified quantitative indicator for subsequent risk level determination.

[0077] It should be noted that TTC (Time to Collision) and DDT (Shortest Braking Distance) jointly define collision risk through the spatial and temporal dimensions: When TTC ≤ DDT, it means that the time required for the vehicle to reach a collision at its current speed is less than or equal to the time required to fully brake to a stop. This indicates that the collision cannot be avoided with existing braking capabilities alone, and there is a real collision risk. Conversely, if TTC > DDT, the vehicle is deemed to have sufficient time and distance to avoid the accident through braking or steering.

[0078] To ensure driving safety, high-risk scenarios must be handled first to ensure that strong collision risks within a short distance are intervened in a timely manner. By utilizing a large number of simulation tests such as scenario libraries, the weight combination of risk assessment indicators such as TTC (time remaining to collision) and DDT (shortest braking distance) is verified to ensure the rationality of the weight setting. Based on historical experience, typical reference values ​​are given, where when TTC = 0.7 and DDT = 0.2, the false trigger rate is 1.2% and the intervention effectiveness reaches 85%; when TTC = 0.6 and DDT = 0.3, the false trigger rate is reduced to 0.8% and the intervention effectiveness is increased to 92%. In actual application, the weights can be recalibrated according to specific circumstances.

[0079] The 0.1 weighting of HMI interaction frequency reflects special considerations for driver attention. Even if the DDT shows ample braking space, if the driver frequently operates the interactive system (such as high-frequency touch screens and voice command input), it will be determined that the degree of distraction is sufficient to increase the risk level. At this time, the risk quantification value is amplified through weighted calculation, and a warning or intervention is forcibly triggered to prompt the driver to take over the vehicle, avoiding the TTC being quickly shortened to within the danger threshold due to distraction.

[0080] Step B2: comparing the risk quantification data with the first threshold to obtain a first comparison result.

[0081] Specifically, the obtained risk quantification data (risk quantification data value) is numerically compared with a preset first threshold (such as 0.3, representing a mild risk critical value).

[0082] The comparison logic is: if the risk quantification data is less than the first threshold, it is judged as "risk is controllable"; if the risk quantification data value is greater than or equal to the first threshold, it is judged as "there is a security risk".

[0083] The comparison process is executed in real time by the domain controller, and the Boolean value result (such as the binary signal of "yes / no trigger intervention") is output through the logic judgment module, providing a direct basis for active safety intervention decisions.

[0084] Step B3, determining whether to trigger active safety intervention for the vehicle during driving based on the first comparison result, wherein, if the first comparison result is that the risk quantification data is less than the first threshold, then no active safety intervention is triggered for the vehicle during driving; or, if the first comparison result is that the risk quantification data is greater than or equal to the first threshold, then active safety intervention is triggered for the vehicle during driving.

[0085] Specifically, if the risk quantification data is less than the first threshold, it indicates that the current driving risk is within a controllable range, and the driver's condition and vehicle environment do not pose a significant safety threat. Therefore, active safety intervention will not be triggered, and only routine monitoring will be maintained (such as continuing to collect sensor data and updating risk models).

[0086] If the risk quantification data is ≥ the first threshold: it is determined that there is a safety risk that requires intervention (such as driver distraction leading to escalated risk, or the vehicle is approaching a critical collision state). At this time, the active safety intervention process is triggered. Further combined with the driving status data (such as fatigue, abnormal posture) and the preset intervention strategy (sound and light reminders, torque limitation), the corresponding safety control operations are performed (such as activating sound and light alarms, limiting motor output power) to reduce the risk of collision or remind the driver to take over the vehicle.

[0087] Step S103: If active safety intervention is triggered for the vehicle, a target intervention strategy is obtained based on the driving status and / or driving risk data.

[0088] In the embodiment of the present application, obtaining target intervention strategies based on driving status and / or driving risk data includes the following situations:

[0089] ① If the driving state is an unsafe driving state, and / or the risk quantification data is less than the second threshold and greater than or equal to the first threshold, the target intervention strategy is determined to be the first intervention strategy, where the risk quantification data is obtained by weighted summation of the remaining collision time, the shortest braking distance, and the number of interactions.

[0090] It should be noted that the first intervention strategy is triggered when the driver is identified as being in an unsafe driving state (such as distraction, fatigue, or abnormal expression), or when the risk quantification data (calculated by the weighted calculation of the remaining collision time TTC, the shortest braking distance DDT, and the number of interactions, with the formula of 0.6×TTC+0.3×DDT+0.1×number of interactions) is between the first threshold (such as 0.3) and the second threshold (such as 0.7). For example, if the driver's eyes are away from the road for more than 3 seconds (unsafe state), or the risk quantification data is 0.5 (between 0.3-0.7), the risk level is determined to be mild. At this time, the first intervention strategy (Level 1) is activated, such as instrument panel audio and video reminders and seat vibration warnings, to prompt the driver to pay attention, while continuously monitoring data to evaluate the effectiveness of the intervention.

[0091] ② If the remaining time to collision in the driving risk data is less than the preset time and the driving state is an unsafe driving state, the target intervention strategy is determined to be the second intervention strategy.

[0092] It should be noted that if the remaining time to collision (TTC) in the driving risk data is less than a preset time (such as 2 seconds), indicating that the collision risk has entered the emergency stage and the driver is also in an unsafe driving state (such as fatigue resulting in no takeover action), the second intervention strategy will be directly triggered. For example, if the TTC is less than 2 seconds and the driver does not respond to the first-stage audio and visual reminder (determined as no takeover action), the conventional warning is deemed to have failed and the second intervention strategy (Level 2) is immediately activated, such as automatically applying partial braking and adjusting steering power to correct the driving path, reducing the collision speed or avoiding path conflict through physical intervention, while continuing to issue a takeover request to the driver.

[0093] ③ If the risk quantification data is greater than or equal to the second threshold, and / or the first intervention strategy or the second intervention strategy is not effective after execution, the targeted intervention strategy is determined to be the third intervention strategy.

[0094] It should be noted that when the risk quantification data is greater than or equal to the second threshold (such as 0.7), it indicates that the collision risk is extremely high (such as extremely short TTC and insufficient DDT), and / or the risk does not decrease after executing the first and second intervention strategies (such as the risk quantification data is still ≥0.7), it is determined that the highest level of intervention must be taken, triggering the third intervention strategy (Level 3).

[0095] For example, if the risk quantification data reaches 0.8 (higher than 0.7), or the first-stage sound and light reminders and the second-stage automatic braking fail to effectively improve TTC or reduce the risk quantification data, the third intervention strategy (Level 3) will be activated, such as full-force emergency braking, active steering avoidance and collision loss analysis.

[0096] Step S104: executing a safety intervention operation corresponding to the target intervention strategy on the vehicle.

[0097] In the embodiment of the present application, when executing the safety intervention operation corresponding to the target intervention strategy on the vehicle, differentiated measures are implemented according to the risk level: when it is determined to be a mild risk, that is, the driving state is abnormal or the risk quantification data is in a lower threshold range, the driver is urged to concentrate through sound and light warnings on the instrument panel, seat vibration, voice reminders, etc. If the remaining time to collision is too short and the driver does not respond to the low-level intervention, it will automatically intervene to apply partial braking force, adjust the steering assist, and limit the motor power. When the risk is extremely high or the previous intervention is ineffective, full-force emergency braking is initiated, and collision loss analysis is performed in combination with path planning, so as to achieve full-process safety protection from early warning to active protection.

[0098] The embodiments of the present application effectively overcome the deficiencies of the existing technology through multi-dimensional data collection and dynamic response mechanisms. Compared with the traditional method that relies on visual perception and machine learning models, the present application simultaneously detects the driver's driving status and vehicle driving risk data to avoid the misjudgment limitations of a single perception. For example, it can assist in judging potential risks through abnormal conditions such as driver distraction and fatigue. Secondly, by analyzing driving risk data in real time, it can determine whether to trigger active safety intervention, change the passive situation of traditional algorithms that focus on post-collision protection, and realize real-time monitoring of driving dangers. According to the driving status and driving risk data, the target intervention strategy is matched, and corresponding safety operations are performed to form a complete closed loop from detection, analysis to active intervention, actively reducing risks before an accident occurs, and making up for the lack of active prevention mechanisms in traditional technologies.

[0099] Figure 2 is a flow chart of a vehicle control method according to an embodiment of the present invention. Figure 2 As shown, the process includes the following steps:

[0100] Step S201 : During the driving of the vehicle, the driver's driving state is detected and the driving risk data of the vehicle is obtained.

[0101] In an embodiment of the present application, multiple types of sensors are used to detect the driver's status and vehicle driving risk data in real time during vehicle driving. For example, a seat pressure sensor array monitors body pressure distribution to determine whether the driving posture is normal, a steering wheel bioelectric sensor collects physiological signals such as heart rate, and an eye tracking camera identifies the focus of vision and blinking frequency to complete driving status detection. A millimeter-wave radar on the roof detects the distance to the target object in front with a driving conflict with an accuracy of ±10cm. The camera identifies lane lines and traffic signs, and combines the three-dimensional environment model constructed by the laser radar to calculate parameters such as the remaining time to collision (TTC) between the vehicle and the target object and the shortest braking distance (DDT) required for emergency braking from the current speed. The number of HMI interactions per unit time is counted to comprehensively form driving risk data.

[0102] Step S202: Analyze the driving risk data to determine whether active safety intervention is triggered for the vehicle during driving.

[0103] In an embodiment of the present application, the domain controller analyzes the acquired driving risk data, calculates the risk quantification value based on a preset risk assessment formula (risk quantification value = 0.6×TTC+0.3×DDT+0.1×number of interactions), and compares it with a first threshold (such as 0.3) and a second threshold (such as 0.7).

[0104] If the risk quantification value is ≥ the first threshold, or the remaining time to collision (TTC) is less than the preset emergency time (such as 2 seconds), it is determined that active safety intervention needs to be triggered; if the risk quantification value is less than the first threshold and there is no emergency time risk, regular monitoring is maintained.

[0105] Step S203: If active safety intervention is triggered for the vehicle, a target intervention strategy is obtained based on the driving status and / or driving risk data.

[0106] In an embodiment of the present application, after determining that active safety intervention is triggered, the target intervention strategy is matched according to the specific combination of the driving state (such as distraction or fatigue) and the driving risk data: if the driving state is an unsafe state (such as distraction), or the risk quantification data is between the first threshold and the second threshold (such as 0.3≤risk quantification data<0.7), the first intervention strategy is hit; if TTC<preset time and the driving state is abnormal, the second intervention strategy is hit; if the risk quantification data ≥the second threshold or the previous intervention is not effective, the third intervention strategy is hit.

[0107] Step S204: If the target intervention strategy is the first intervention strategy, intervention signals are sent to the sound and light alarm device, seat vibration device and display device located inside the vehicle respectively, so as to control the sound and light alarm device to perform the sound and light alarm operation, control the seat vibration device to perform the vibration operation and control the display device to display the warning prompt information through the intervention signal.

[0108] In the embodiment of the present application, when it is determined that the first intervention strategy is hit, the domain controller acts as a central control unit and sends a standardized digital signal to the execution device inside the vehicle via the CAN bus. The specific implementation process is as follows:

[0109] The domain controller sends a PWM (pulse width modulation) signal to the instrument panel lighting control module, driving the red warning light to flash at a frequency of 1Hz; at the same time, it sends instructions to the audio controller to trigger the buzzer to emit an 80dB, 2000Hz high-frequency prompt sound. The light and sound signals attract the driver's attention through both visual and auditory channels.

[0110] The domain controller generates a periodic pulse signal (such as a square wave with a duty cycle of 50% and a period of 2 seconds), which is transmitted to the vibration motor built into the seat through the drive circuit. This causes the seat back or seat cushion to vibrate 3 times every 2 seconds (each vibration lasting 200 milliseconds), generating tactile feedback and enhancing the driver's physical perception.

[0111] The domain controller sends graphic data frames to the instrument panel screen via the vehicle's Ethernet or CAN bus. These frames contain predefined warning icons (such as a yellow exclamation mark) and the text message "Please pay attention." The display module parses the data and highlights it in the center of the screen with a 24pt font size and red color to ensure clear visibility within the driver's field of view.

[0112] Various devices parse data packets on the CAN bus through their respective signal receiving modules (such as lighting controllers, motor drivers, and display controllers), identify instruction types based on identifiers (IDs), and synchronously execute corresponding warning operations, forming a "visual-auditory-tactile" multimodal reminder combination, effectively improving the driver's response efficiency to unsafe conditions and prompting them to correct their driving behavior in a timely manner.

[0113] When the first intervention strategy is triggered in the embodiment of the present application, multimodal warnings such as sound and light alarms, seat vibrations, and screen prompts are used to quickly awaken the driver's attention with three-dimensional stimulation of visual flashes, auditory buzzes, and tactile vibrations, prompting them to correct their driving behavior in a timely manner, thereby effectively reducing the probability of accidents at the initial stage of risk.

[0114] Figure 3 is a flow chart of a vehicle control method according to an embodiment of the present invention. Figure 3 As shown, the process includes the following steps:

[0115] Step S301: During vehicle driving, the driver's driving state is detected and driving risk data of the vehicle is obtained.

[0116] In an embodiment of the present application, multiple types of sensors are used to detect the driver's status and vehicle driving risk data in real time during vehicle driving. For example, a seat pressure sensor array monitors body pressure distribution to determine whether the driving posture is normal, a steering wheel bioelectric sensor collects physiological signals such as heart rate, and an eye tracking camera identifies the focus of vision and blinking frequency to complete driving status detection. A millimeter-wave radar on the roof detects the distance to the target object in front with a driving conflict with an accuracy of ±10cm. The camera identifies lane lines and traffic signs, and combines the three-dimensional environment model constructed by the laser radar to calculate parameters such as the remaining time to collision (TTC) between the vehicle and the target object and the shortest braking distance (DDT) required for emergency braking from the current speed. The number of HMI interactions per unit time is counted to comprehensively form driving risk data.

[0117] Step S302: Analyze the driving risk data to determine whether active safety intervention is triggered for the vehicle during driving.

[0118] In an embodiment of the present application, the domain controller analyzes the acquired driving risk data, calculates the risk quantification value based on a preset risk assessment formula (risk quantification value = 0.6×TTC+0.3×DDT+0.1×number of interactions), and compares it with a first threshold (such as 0.3) and a second threshold (such as 0.7).

[0119] If the risk quantification value is ≥ the first threshold, or the remaining time to collision (TTC) is less than the preset emergency time (such as 2 seconds), it is determined that active safety intervention needs to be triggered; if the risk quantification value is less than the first threshold and there is no emergency time risk, regular monitoring is maintained.

[0120] Step S303: If active safety intervention is triggered for the vehicle, a target intervention strategy is obtained based on the driving status and / or driving risk data.

[0121] In an embodiment of the present application, after determining that active safety intervention is triggered, the target intervention strategy is matched according to the specific combination of the driving state (such as distraction or fatigue) and the driving risk data: if the driving state is an unsafe state (such as distraction), or the risk quantification data is between the first threshold and the second threshold (such as 0.3≤risk quantification data<0.7), the first intervention strategy is hit; if TTC<preset time and the driving state is abnormal, the second intervention strategy is hit; if the risk quantification data ≥the second threshold or the previous intervention is not effective, the third intervention strategy is hit.

[0122] Step S304 : If the target intervention strategy is the second intervention strategy, the power output of the vehicle is limited, and the rotation angle of the vehicle steering wheel is limited to a set angle range.

[0123] In this embodiment of the present application, if the second intervention strategy is triggered, the domain controller sends control instructions to the vehicle's powertrain and steering systems via the CAN bus. Regarding power output, the domain controller sends a signal to the motor controller to limit the motor's torque output, for example, by reducing the power output by 40% and thus reducing the vehicle's speed. Regarding steering control, the domain controller sends instructions to the electric power steering system to limit the steering wheel's rotation angle to a set range (e.g., ±15°) to prevent the driver from losing control of the vehicle due to panicked operation. At the same time, the domain controller automatically applies partial braking force to keep the vehicle traveling at a safe speed and trajectory, reducing the risk of collision.

[0124] In the embodiment of the present application, when the second intervention strategy is triggered, in terms of power output, the domain controller sends a power limit signal to the motor controller or the engine to reduce the vehicle's power output, such as reducing the motor torque by 40% or reducing the amount of fuel injection, so that the vehicle speed decreases steadily to avoid uncontrolled collisions caused by excessive speed. In terms of steering control, the domain controller issues instructions to the electric power steering system to limit the steering wheel rotation angle to a set range (such as ±15°) to prevent the driver from making large turns due to panic or misoperation, causing the vehicle to roll over or collide with other obstacles. These two physical restriction measures work together. When the driver fails to respond correctly in time, technical means are used to force the vehicle speed to be reduced and the vehicle's driving trajectory to be stabilized, thereby minimizing the possibility of accidents.

[0125] For example, consider a vehicle traveling at 60 km / h on an urban road. A pedestrian suddenly appears ahead, the vehicle detects a time to collision (TTC) of less than 2 seconds, and the driver fails to take effective braking or evasive action. This triggers the second intervention strategy. The domain controller then sends a command to the powertrain via the CAN bus, reducing the motor's power output by 50%, bringing the vehicle speed down to 30 km / h within 1.5 seconds and mitigating the impact. Simultaneously, the domain controller instructs the electric power steering system to limit the steering wheel angle to ±10°, preventing the driver from panicking and causing the vehicle to roll over or collide with a vehicle in the adjacent lane. Ultimately, the vehicle makes minor contact with the pedestrian at a lower speed, significantly reducing injury to the pedestrian and the likelihood of a secondary accident compared to a high-speed, uncontrolled collision.

[0126] Figure 4 is a flow chart of a vehicle control method according to an embodiment of the present invention. Figure 4 As shown, the process includes the following steps:

[0127] Step S401: During the driving of the vehicle, the driver's driving state is detected and the driving risk data of the vehicle is obtained.

[0128] In an embodiment of the present application, multiple types of sensors are used to detect the driver's status and vehicle driving risk data in real time during vehicle driving. For example, a seat pressure sensor array monitors body pressure distribution to determine whether the driving posture is normal, a steering wheel bioelectric sensor collects physiological signals such as heart rate, and an eye tracking camera identifies the focus of vision and blinking frequency to complete driving status detection. A millimeter-wave radar on the roof detects the distance to the target object in front with a driving conflict with an accuracy of ±10cm. The camera identifies lane lines and traffic signs, and combines the three-dimensional environment model constructed by the laser radar to calculate parameters such as the remaining time to collision (TTC) between the vehicle and the target object and the shortest braking distance (DDT) required for emergency braking from the current speed. The number of HMI interactions per unit time is counted to comprehensively form driving risk data.

[0129] Step S202: Analyze the driving risk data to determine whether active safety intervention is triggered for the vehicle during driving.

[0130] In an embodiment of the present application, the domain controller analyzes the acquired driving risk data, calculates the risk quantification value based on a preset risk assessment formula (risk quantification value = 0.6×TTC+0.3×DDT+0.1×number of interactions), and compares it with a first threshold (such as 0.3) and a second threshold (such as 0.7).

[0131] If the risk quantification value is ≥ the first threshold, or the remaining time to collision (TTC) is less than the preset emergency time (such as 2 seconds), it is determined that active safety intervention needs to be triggered; if the risk quantification value is less than the first threshold and there is no emergency time risk, regular monitoring is maintained.

[0132] Step S403: If active safety intervention is triggered for the vehicle, a target intervention strategy is obtained based on the driving status and / or driving risk data.

[0133] In an embodiment of the present application, after determining that active safety intervention is triggered, the target intervention strategy is matched according to the specific combination of the driving state (such as distraction or fatigue) and the driving risk data: if the driving state is an unsafe state (such as distraction), or the risk quantification data is between the first threshold and the second threshold (such as 0.3≤risk quantification data<0.7), the first intervention strategy is hit; if TTC<preset time and the driving state is abnormal, the second intervention strategy is hit; if the risk quantification data ≥the second threshold or the previous intervention is not effective, the third intervention strategy is hit.

[0134] Step S404 : If the target intervention strategy is the third intervention strategy, the vehicle is controlled to perform emergency braking, and the number and type of target objects that conflict with the vehicle are obtained.

[0135] In the embodiment of the present application, the vehicle is first controlled to perform emergency braking. Then, through the fusion perception of the lidar and camera on the roof, a point cloud clustering algorithm is used to identify target objects (such as pedestrians, vehicles, and non-motor vehicles) that are in conflict with the vehicle and count the number of objects. At the same time, the target objects are classified based on a convolutional neural network (CNN) to determine their types (such as small cars, heavy trucks, and pedestrians). For example, if two vehicles decelerating in the same direction and one pedestrian crossing the road are detected within 30 meters ahead, the system records the number of objects as 3 and marks the object types as "vehicle" and "pedestrian" respectively.

[0136] Step S405 : planning a safe driving path for the vehicle based on the number and type of objects.

[0137] In an embodiment of the present application, planning a safe driving path for a vehicle based on the number and type of objects includes: planning at least one candidate driving path based on the number of objects; obtaining object weights corresponding to the object types, where the object weights are used to indicate the degree of protection provided by different types of objects in the event of a driving conflict with the vehicle; calculating a risk value for each candidate driving path using the number of objects and the object weights; and selecting the candidate driving path with the lowest risk value as the safe driving path for the vehicle.

[0138] Specifically, through lidar point cloud clustering and camera image recognition algorithms, the number of target objects that may conflict with the vehicle (such as two pedestrians and one stationary vehicle within 30 meters ahead) is detected in real time. Combined with high-precision map lane information and vehicle dynamic boundaries (minimum turning radius, maximum lateral acceleration), a fast search random tree algorithm is used to generate multiple candidate driving paths.

[0139] A matrix of preset object types and weight coefficients is set. When the sensor identifies the target object type (such as pedestrians, passengers, and property), it automatically matches the corresponding weight. The object weight setting rules are as follows: pedestrians > passengers > property, and the weight of children is doubled.

[0140] For example, a pedestrian has a weight of 0.7. A passenger has a weight of 0.5. Property (guardrails, roadside facilities) has a weight of 0.3. If a child is detected, the child protection rule is triggered, multiplying the pedestrian weight by 2 (i.e., 0.7 × 2 = 1.4). If the collision targets are a child and a guardrail, the corresponding weights are 1.4 and 0.3, respectively.

[0141] For each candidate path, all conflicting objects on its predicted trajectory are traversed, and the total damage risk value is calculated based on the number of objects and the weight coefficient. The formula is: Risk value = ∑(number of objects × object weight × loss baseline value). The loss baseline value can be set to 1 by default. For example:

[0142] Path A: A collision with a pedestrian is predicted, with a risk value of 1 × 0.7 × 1 = 0.7.

[0143] Path B: Collision with two guardrails is predicted, with a risk value of 2 × 0.3 × 1 = 0.6.

[0144] Path B: A pedestrian is predicted to collide with two guardrails. The risk value is 1 × 0.7 × 1 + 2 × 0.3 × 1 = 1.3.

[0145] Compare the risk values ​​of all candidate paths and select the path with the lowest risk value as the final execution plan. For example, in the above example, path B < path A < path C, so path B is selected as the safest driving path.

[0146] The path coordinate sequence is then converted into control instructions such as steering angle, brake pressure, and motor torque via the CAN bus and sent to the electronic mechanical braking system (response time <100ms) and electric power steering system. By limiting the steering wheel angle (e.g., ±20°) and precise braking, the vehicle is ensured to travel along the selected path. At the same time, environmental changes are continuously monitored, and path planning is retriggered if a new conflicting object suddenly appears.

[0147] It should be noted that object weights can be set according to different scenarios. Object weights can be adaptively adjusted based on time, space, driver status and environmental factors. When a pedestrian suddenly crosses the road within 0-2 seconds, the pedestrian risk weight is temporarily increased to 0.85 times the baseline value and recovers after 3 seconds; in special areas such as crosswalks and schools, the pedestrian weight is directly increased to 0.85; if the biological signal monitors the driver's sudden illness, the passenger weight is increased to 0.7, giving priority to ensuring safety in the car; and when the lidar recognizes icy and snowy roads, the pedestrian weight is reduced to 0.65 due to limited braking, balancing collision avoidance and driving stability, so as to ensure that the active safety strategy meets the actual scenario needs.

[0148] Step S406: Control the vehicle to travel along a safe driving path.

[0149] In the embodiment of the present application, the domain controller converts the planned safe driving path into a control command and sends it to the vehicle actuator via the CAN bus. The electric power steering (EPS) system receives the steering angle command and automatically adjusts the steering wheel angle; the power system adjusts the motor torque or engine output power according to the path requirements to control the vehicle speed; and the braking system applies braking force as needed to ensure that the vehicle travels smoothly along the planned path. For example, when the path is planned to change lanes to the left, the domain controller controls the EPS to turn the steering wheel 12° and reduces the motor power to reduce the vehicle speed from 60km / h to 50km / h, cooperating with ESP to stabilize the vehicle body posture until the vehicle is completely in a safe lane.

[0150] It should be noted that when the third intervention strategy is triggered, it means that the risk quantification data has reached or exceeded the threshold of 0.7, indicating that the vehicle's driving condition is approaching or entering an emergency state where collision cannot be completely avoided. At this time, relying solely on conventional braking or warnings is insufficient to eliminate the danger. In this case, the system obtains the number and type of target objects in conflict with the vehicle, plans at least one candidate driving path based on the number of objects, and comprehensively considers the protection priority of different objects in the collision according to a preset object weighting system. By calculating the risk value of each candidate driving path, using the number of objects, object weights, and damage baseline as parameters, the potential collision damage level of each path is evaluated. Ultimately, the candidate driving path with the lowest risk value is selected as the vehicle's safe driving path. In the extreme case of unavoidable collision, this method minimizes damage to high-priority protected objects, balances various risks, and achieves emergency optimization and safety control of the vehicle's driving path.

[0151] When the third intervention strategy is triggered, the embodiment of the present application obtains the number and type of conflicting target objects, plans candidate driving paths based on the object weights and calculates the risk value, selects the path with the lowest risk, gives priority to protecting high-weight objects (such as pedestrians), and takes into account the safety of the vehicle itself and other objects; finally, the vehicle is controlled to travel along a safe path. In extremely dangerous situations, through intelligent path planning and active control, accident damage is minimized to the greatest extent, thereby improving driving safety and reliability in complex traffic scenarios.

[0152] This embodiment also provides a vehicle control device for implementing the above-described embodiments and preferred embodiments. Details already described will not be repeated. As used below, the term "module" may refer to a combination of software and / or hardware that implements a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, implementation using hardware, or a combination of software and hardware, is also possible and contemplated.

[0153] This embodiment provides a vehicle control device, such as Figure 5 Shown, including:

[0154] The detection module 501 is used to detect the driver's driving state and obtain the vehicle's driving risk data during the vehicle's driving process;

[0155] An analysis module 502 is used to analyze driving risk data and determine whether to trigger active safety intervention for the vehicle during driving;

[0156] An acquisition module 503 is configured to acquire a target intervention strategy for the driving state and / or driving risk data if active safety intervention is triggered for the vehicle;

[0157] The execution module 504 is configured to execute the safety intervention operation corresponding to the target intervention strategy on the vehicle.

[0158] In an embodiment of the present application, the detection module 501 is used to detect a target object that is in driving conflict with the vehicle and obtain the remaining collision time between the vehicle and the target object; obtain the shortest braking distance required for the vehicle to be completely braked from the current driving speed to a standstill; obtain the number of interactions between the driver and the vehicle's interactive system per unit time; and use the remaining collision time, the shortest braking distance, and the number of interactions as driving risk data.

[0159] In an embodiment of the present application, the analysis module 502 is used to perform a weighted sum based on the remaining collision time, the shortest braking distance, and the number of interactions to obtain risk quantification data; compare the risk quantification data and a first threshold to obtain a first comparison result; determine whether to trigger active safety intervention for the vehicle during vehicle driving based on the first comparison result, wherein, if the first comparison result is that the risk quantification data is less than the first threshold, then active safety intervention is not triggered for the vehicle during vehicle driving; or, if the first comparison result is that the risk quantification data is greater than or equal to the first threshold, then active safety intervention is triggered for the vehicle during vehicle driving.

[0160] In an embodiment of the present application, the acquisition module 503 is used to determine that the target intervention strategy that is hit is the first intervention strategy if the driving state is an unsafe driving state, and / or the risk quantification data is less than the second threshold and greater than or equal to the first threshold, wherein the risk quantification data is obtained by weighted summation of the remaining collision time, the shortest braking distance, and the number of interactions; or, if the remaining collision time in the driving risk data is less than the preset time, and the driving state is an unsafe driving state, then the target intervention strategy that is hit is determined to be the second intervention strategy; or, if the risk quantification data is greater than or equal to the second threshold, and / or, the first intervention strategy or the second intervention strategy does not take effect after execution, then the target intervention strategy that is hit is determined to be the third intervention strategy.

[0161] In an embodiment of the present application, the execution module 504 is used to send intervention signals to the sound and light alarm device, seat vibration device and display device located inside the vehicle respectively if the target intervention strategy is the first intervention strategy, so as to control the sound and light alarm device to perform the sound and light alarm operation, control the seat vibration device to perform the vibration operation and control the display device to display the warning prompt information through the intervention signal.

[0162] In the embodiment of the present application, the execution module 504 is configured to limit the power output of the vehicle and limit the rotation angle of the vehicle steering wheel to a set angle range if the target intervention strategy is the second intervention strategy.

[0163] In an embodiment of the present application, the execution module 504 is used to control the vehicle to perform emergency braking if the target intervention strategy is the third intervention strategy, and obtain the number and type of target objects that have a driving conflict with the vehicle; plan a safe driving path for the vehicle based on the number and type of objects; and control the vehicle to drive along the safe driving path.

[0164] In an embodiment of the present application, the execution module 504 is used to plan at least one candidate driving path based on the number of objects; obtain the object weight corresponding to the object type, wherein the object weight is used to indicate the degree of protection of different types of objects when a driving conflict occurs with the vehicle; calculate the risk value of each candidate driving path using the number of objects and the object weight; and select the candidate driving path with the smallest risk value as the safe driving path for the vehicle.

[0165] An embodiment of the present application provides a vehicle, comprising: a memory and a controller, the memory and the controller being communicatively connected to each other, the memory storing computer instructions, and the controller executing the above method by executing the computer instructions.

[0166] See also Figure 6 , Figure 6 is a structural diagram of an electronic device provided by an optional embodiment of the present invention, such as Figure 6 As shown, the electronic device includes: one or more processors 10, a memory 20, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. The various components utilize different buses to communicate with each other and can be installed on a common mainboard or installed in other ways as needed. The processor can process instructions executed in the electronic device, including instructions stored in or on the memory to display graphical information of a GUI on an external input / output device (such as a display device coupled to the interface). In some optional embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Similarly, multiple electronic devices can be connected, and each device provides some necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system).

[0167] The processor 10 may be a central processing unit, a network processor, or a combination thereof. The processor 10 may further include a hardware chip. The hardware chip may be an application-specific integrated circuit, a programmable logic device, or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable gate array, a general purpose array logic, or any combination thereof.

[0168] The memory 20 stores instructions that can be executed by at least one processor 10, so that the at least one processor 10 executes the method shown in the above embodiment.

[0169] The memory 20 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created based on the use of an electronic device presented by a small program landing page, etc. In addition, the memory 20 may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some optional embodiments, the memory 20 may optionally include a memory remotely located relative to the processor 10, and these remote memories may be connected to the electronic device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

[0170] The memory 20 may include a volatile memory, such as a random access memory; the memory may also include a non-volatile memory, such as a flash memory, a hard disk or a solid-state drive; the memory 20 may also include a combination of the above types of memory.

[0171] The embodiment of the present invention also provides a computer-readable storage medium. The above-mentioned method according to the embodiment of the present invention can be implemented in hardware, firmware, or implemented as a computer code that can be recorded in a storage medium, or implemented as a computer code that is originally stored in a remote storage medium or a non-temporary machine-readable storage medium and downloaded through a network and will be stored in a local storage medium, so that the method described herein can be stored in such software processing on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only storage memory, a random access memory, a flash memory, a hard disk or a solid-state drive, etc.; further, the storage medium can also include a combination of the above-mentioned types of memory. It can be understood that a computer, a processor, a microprocessor controller or programmable hardware includes a storage component that can store or receive software or computer code. When the software or computer code is accessed and executed by a computer, a processor or hardware, the method shown in the above embodiment is implemented.

[0172] Although the embodiments of the present invention have been described with reference to the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present invention. Such modifications and variations are all within the scope defined by the appended claims.

Claims

1. A vehicle control method, characterized in that: The method comprises: During vehicle driving, detect the driver's driving status and obtain vehicle driving risk data; Analyzing the driving risk data to determine whether to trigger active safety intervention for the vehicle during driving; If active safety intervention is triggered for the vehicle, obtaining a target intervention strategy matched with the driving state and / or the driving risk data; Executing a safety intervention operation corresponding to the target intervention strategy on the vehicle.

2. The method according to claim 1, characterized in that The obtaining of the vehicle's driving risk data includes: Detecting a target object that is in conflict with the vehicle, and obtaining a remaining collision time between the vehicle and the target object; Obtaining the shortest braking distance required for the vehicle to be completely braked from the current speed to a standstill; Obtaining the number of interactions between the driver and the vehicle's interactive system within a unit time; The remaining time to collision, the shortest braking distance, and the number of interactions are used as the driving risk data.

3. The method according to claim 2, characterized in that The analyzing the driving risk data to determine whether to trigger active safety intervention for the vehicle during driving includes: Performing a weighted sum based on the remaining collision time, the shortest braking distance, and the number of interactions to obtain risk quantification data; Comparing the risk quantification data with a first threshold to obtain a first comparison result; Determine whether to trigger active safety intervention for the vehicle during driving based on the first comparison result, wherein, if the first comparison result is that the risk quantification data is less than the first threshold, active safety intervention is not triggered for the vehicle during driving; or, if the first comparison result is that the risk quantification data is greater than or equal to the first threshold, active safety intervention is triggered for the vehicle during driving.

4. The method according to claim 2, characterized in that The obtaining of the target intervention strategy hit by the driving state and / or the driving risk data includes: If the driving state is an unsafe driving state, and / or the risk quantification data is less than a second threshold and greater than or equal to a first threshold, determining that the targeted intervention strategy is the first intervention strategy, wherein the risk quantification data is obtained by weighted summation of the remaining time to collision, the shortest braking distance, and the number of interactions; or If the remaining time to collision in the driving risk data is less than the preset time and the driving state is an unsafe driving state, determining that the target intervention strategy to be hit is the second intervention strategy; or, If the risk quantification data is greater than or equal to the second threshold, and / or the first intervention strategy or the second intervention strategy does not take effect after execution, it is determined that the hit target intervention strategy is the third intervention strategy.

5. The method according to claim 4, characterized in that The executing the safety intervention operation corresponding to the target intervention strategy on the vehicle includes: If the target intervention strategy is the first intervention strategy, intervention signals are sent to the sound and light alarm device, seat vibration device and display device located inside the vehicle respectively, so as to control the sound and light alarm device to perform sound and light alarm operations, control the seat vibration device to perform vibration operations and control the display device to display warning prompt information through the intervention signal.

6. The method according to claim 4, characterized in that The executing the safety intervention operation corresponding to the target intervention strategy on the vehicle includes: If the target intervention strategy is the second intervention strategy, the power output of the vehicle is limited, and the rotation angle of the vehicle steering wheel is limited to a set angle range.

7. The method according to claim 4, characterized in that The executing the safety intervention operation corresponding to the target intervention strategy on the vehicle includes: If the target intervention strategy is the third intervention strategy, controlling the vehicle to perform emergency braking, and obtaining the number and type of target objects that have a driving conflict with the vehicle; Planning a safe driving path for the vehicle based on the number of objects and the types of objects; The vehicle is controlled to travel along the safe driving path.

8. The method according to claim 7, characterized in that Planning a safe driving path for the vehicle based on the number of objects and the types of objects includes: planning at least one candidate driving path based on the number of objects; Obtaining an object weight corresponding to the object type, wherein the object weight is used to indicate a degree of protection of different types of objects when a driving conflict occurs with the vehicle; Calculating a risk value for each candidate driving path using the number of objects and the object weights; The candidate driving path with the smallest risk value is used as the safe driving path for the vehicle.

9. A vehicle control device, characterized in that: The device comprises: The detection module is used to detect the driver's driving status and obtain the vehicle's driving risk data during the vehicle's driving process; An analysis module, configured to analyze the driving risk data and determine whether to trigger active safety intervention for the vehicle during driving; an acquisition module, configured to acquire a target intervention strategy hit by the driving state and / or the driving risk data if active safety intervention is triggered for the vehicle; An execution module is used to execute the safety intervention operation corresponding to the target intervention strategy on the vehicle.

10. A vehicle, characterized in that: include: A memory and a controller, wherein the memory and the controller are communicatively connected to each other, the memory stores computer instructions, and the controller executes the method according to any one of claims 1 to 8 by executing the computer instructions.

Images