Vehicle control method and vehicle

Abstract

The embodiment of the application provides a vehicle control method, applied to a target vehicle, the method comprising: receiving front vehicle running data and front vehicle environment perception data of a front vehicle, wherein the front vehicle comprises at least one networked communication vehicle located in front of a driving path of the target vehicle; performing screening processing on the front vehicle running data and the front vehicle environment perception data according to self-vehicle running data of the target vehicle and road topological information of the driving path, to obtain a front vehicle effective data set; predicting environment risk information of a front road section based on the front vehicle running data, the self-vehicle running data and the front vehicle effective data set; and generating joint control instructions of the target vehicle according to the environment risk information, wherein the joint control instructions are used for driving control of multiple execution mechanisms of the target vehicle. The application solves the technical problems of the prior art, such as lagging environment risk perception, poor driving safety and comfort caused by limited single-vehicle perception range.

Description

Technical Field

[0001] This application relates to the field of vehicle technology, and more specifically, to a vehicle control method and a vehicle. Background Technology

[0002] With the development of intelligent connected technologies, vehicle-to-everything (V2X) has been applied to traffic incident broadcasting, providing a foundation for vehicle-road cooperation.

[0003] Current vehicle environmental perception mainly relies on single-vehicle sensors, which result in isolated perception and decision-making, limiting the detection range to the vehicle's surroundings and making it impossible to predict actual environmental risks ahead. Especially when entering harsh environments at high speeds, single-vehicle perception exhibits significant lag and lacks accurate prediction and proactive control capabilities based on multi-vehicle collaborative data, leading to response delays and posing safety hazards.

[0004] There is currently no good solution to the above problems. Summary of the Invention

[0005] This application provides a vehicle control method and a vehicle to at least solve the technical problems of existing technologies, such as delayed environmental risk perception and poor driving safety and comfort due to the limited perception range of a single vehicle.

[0006] According to one aspect of the embodiments of this application, a vehicle control method is provided, applied to a target vehicle, comprising: receiving preceding vehicle operation data and preceding vehicle environmental perception data, wherein the preceding vehicle includes at least one connected communication vehicle located ahead of the target vehicle's driving path; filtering the preceding vehicle operation data and preceding vehicle environmental perception data based on the target vehicle's own vehicle operation data and road topology information of the driving path to obtain a set of valid preceding vehicle data; predicting environmental risk information of the preceding road segment based on the preceding vehicle operation data, the target vehicle's own vehicle operation data, and the set of valid preceding vehicle data; and generating joint control instructions for the target vehicle based on the environmental risk information, wherein the joint control instructions are used to drive and control multiple actuators of the target vehicle.

[0007] Furthermore, the preceding vehicle's operating data and the preceding vehicle's environmental perception data are obtained by parsing the wiper status messages broadcast by the preceding vehicle at a target frequency. The target frequency is dynamically adjusted by the preceding vehicle according to the wiper's operating mode. The wiper status messages also encapsulate a message identifier field and a data confidence field.

[0008] Furthermore, the method also includes: in response to receiving an emergency event message broadcast by the vehicle ahead, parsing the emergency event message to obtain updated vehicle ahead operation data and updated vehicle ahead environmental perception data, wherein the emergency event message is generated by the vehicle ahead when it detects that the rate of change of the windshield wiper operating frequency within a preset time period is greater than a preset threshold, and the priority of the emergency event message is higher than the priority of the windshield wiper status message.

[0009] Furthermore, the vehicles ahead include multiple target vehicles. Based on the vehicle's own operation data and road topology information, the preceding vehicle's operation data and preceding vehicle's environmental perception data are filtered and processed to obtain the following effective preceding vehicle data set: extracting the target vehicle's current position, speed, and heading angle from the vehicle's own operation data; traversing multiple target vehicles, for the current target vehicle in the traversal, extracting the current target vehicle's position, speed, heading angle, and message timestamp from the preceding vehicle's operation data corresponding to the current target vehicle; calculating the expected position of the current target vehicle at the current moment based on the vehicle's position, speed, heading angle, and message timestamp; determining whether the expected position is within the effective area corresponding to the driving path based on road topology information, vehicle position, speed, and heading angle; if the expected position is within the effective area, adding the preceding vehicle's operation data and environmental perception data corresponding to the current target vehicle to the effective preceding vehicle data set.

[0010] Furthermore, the environmental risk information includes a precipitation risk coefficient. Based on the preceding vehicle's operating data, the vehicular operating data, and the effective preceding vehicle data set, the predicted environmental risk information includes: determining the wiper operating frequency sequence and precipitation intensity estimation sequence corresponding to the target vehicle based on the effective preceding vehicle data set; determining the relative distance sequence corresponding to the target vehicle based on the preceding vehicle's operating data and the vehicular operating data; normalizing the wiper operating frequency sequence, precipitation intensity estimation sequence, relative distance sequence, and the target vehicle's vehicular environmental parameters to obtain a multi-dimensional feature vector; and inputting the multi-dimensional feature vector into the target prediction model for inference to obtain the precipitation risk coefficient. The target prediction model is a deep neural network model trained using multiple sets of sample data.

[0011] Furthermore, based on environmental risk information, the joint control instructions are generated as follows: based on environmental risk information, the warning level is determined; based on the warning control strategy corresponding to the warning level, multiple sub-control instructions are generated for multiple actuators, including: windshield wiper system, lighting system and air conditioning system.

[0012] Furthermore, the warning control strategy includes: when the warning level is low-risk, a first sub-control command is generated, which controls the wiper system to perform a preparation mode switching action and a wiper blade reset action; when the warning level is medium-risk, a second sub-control command is generated, which controls the wiper system to perform a wiping action at a first wiping frequency determined based on environmental risk information; when the warning level is high-risk, a third, fourth, and fifth sub-control command are generated, whereby the third sub-control command controls the wiper system to switch to a target continuous wiping mode, the second wiping frequency corresponding to the target continuous wiping mode being higher than the first wiping frequency; the fourth sub-control command controls the lighting system to turn on the front and rear fog lights; and the fifth sub-control command controls the air conditioning system to switch to the windshield defroster mode.

[0013] Furthermore, the method also includes: after the target vehicle enters the road section ahead, determining the actual precipitation intensity and recording the corresponding intervention and control information of the target vehicle; determining the model prediction bias based on the actual precipitation intensity and precipitation risk coefficient; constructing feedback sample data based on the intervention and control information and the model prediction bias; and using the feedback sample data to fine-tune the parameters of the target prediction model to update the target prediction model.

[0014] According to another aspect of the embodiments of this application, a vehicle control device is also provided, applied to a target vehicle, comprising: a receiving module, configured to receive preceding vehicle operation data and preceding vehicle environmental perception data, wherein the preceding vehicle includes at least one connected communication vehicle located ahead of the target vehicle's driving path; a filtering module, configured to filter the preceding vehicle operation data and preceding vehicle environmental perception data according to the target vehicle's own vehicle operation data and road topology information of the driving path, to obtain a set of valid preceding vehicle data; a prediction module, configured to predict environmental risk information of the road segment ahead based on the preceding vehicle operation data, the own vehicle operation data, and the set of valid preceding vehicle data; and a generation module, configured to generate joint control instructions for the target vehicle according to the environmental risk information, wherein the joint control instructions are used to drive and control multiple actuators of the target vehicle.

[0015] According to another aspect of the embodiments of this application, a vehicle is also provided, including: a memory storing an executable program; and a processor for running the program, wherein the program executes the methods in various embodiments of this application when it runs.

[0016] According to another aspect of the embodiments of this application, a computer-readable storage medium is also provided, the computer-readable storage medium including a stored executable program, wherein, when the executable program is running, it controls the device where the computer-readable storage medium is located to perform the methods of various embodiments of this application.

[0017] In this embodiment, a vehicle control method is employed. This method receives preceding vehicle operation data and preceding vehicle environmental perception data, where the preceding vehicle includes at least one connected vehicle located ahead of the target vehicle's driving path. Based on the target vehicle's own vehicle operation data and the road topology information of the driving path, the preceding vehicle operation data and preceding vehicle environmental perception data are filtered to obtain a set of effective preceding vehicle data. Based on the preceding vehicle operation data, the target vehicle's own vehicle operation data, and the set of effective preceding vehicle data, environmental risk information for the upcoming road segment is predicted. Based on the environmental risk information, a joint control command for the target vehicle is generated. This joint control command is used to drive and control multiple actuators of the target vehicle. This achieves the goal of using vehicle-to-vehicle communication, combined with preceding vehicle operation data, preceding vehicle environmental perception data, the target vehicle's own vehicle operation data, and road topology information to drive and control the vehicle's actuators in advance to avoid environmental risks on the upcoming road segment. This enhances the real-time performance of environmental risk prediction for the upcoming road segment, thereby improving driving safety and comfort. It also solves the technical problem in existing technologies where the limited perception range of a single vehicle leads to delayed environmental risk perception and poor driving safety and comfort. Attached Figure Description

[0018] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings: Figure 1 This is a flowchart of a vehicle control method according to an embodiment of this application; Figure 2 This is a structural block diagram of a vehicle control system according to an embodiment of this application; Figure 3 This is a schematic flowchart of another vehicle control method according to an embodiment of this application; Figure 4 This is a flowchart of another vehicle control method according to an embodiment of this application; Figure 5 This is a structural block diagram of a vehicle control device according to an embodiment of this application. Detailed Implementation

[0019] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0020] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0021] According to an embodiment of this application, a vehicle control method embodiment is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0022] This embodiment provides a vehicle control method applied to a target vehicle. Figure 1 This is a flowchart of a vehicle control method according to an embodiment of this application, such as... Figure 1 As shown, the process includes the following steps: Step S11: Receive the preceding vehicle's operating data and the preceding vehicle's environmental perception data, wherein the preceding vehicle includes at least one connected communication vehicle located ahead of the target vehicle's driving path. In this embodiment of the application, the target vehicle refers to the main vehicle that executes the vehicle control method of this application, that is, the vehicle that is in the rear position, receives broadcast data from the vehicle in front, and coordinates the control of its own actuators accordingly.

[0023] The target vehicle receives the preceding vehicle's operating data and the preceding vehicle's environmental perception data, wherein the preceding vehicle includes at least one connected communication vehicle located ahead of the target vehicle's driving path.

[0024] Forward vehicle operation data refers to parameter information collected and transmitted by the vehicle in front, reflecting its own vehicle operation status, such as the current working frequency of the windshield wipers, vehicle speed, vehicle position coordinates, heading angle, etc.

[0025] Forward vehicle environment perception data refers to information collected and transmitted by the vehicle in front that reflects the external environment conditions, such as wiper frequency, rainfall data, light intensity, ambient temperature, etc.

[0026] Connected vehicles refer to vehicles that have vehicle-to-vehicle communication capabilities and can exchange data with surrounding vehicles through dedicated short-range communication or cellular vehicle networking technology, such as passenger cars and commercial vehicles equipped with vehicle-to-vehicle (V2V) communication modules.

[0027] It can be seen that the target vehicle listens to and parses the data packets broadcast from the vehicle in front through its on-board communication module, and extracts specific numerical fields containing the operating status of the vehicle in front and the results of environmental perception.

[0028] For example, the target vehicle receives a data packet sent by a smart connected vehicle ahead through a V2V module. After parsing the data packet, it learns that the windshield wiper frequency of the vehicle ahead is 30 times per minute, and the vehicle ahead measures the rainfall intensity of the road section as 5 millimeters per hour.

[0029] As a result, the target vehicle can obtain real-time road precipitation and road condition information beyond visual range, providing basic data support for the subsequent predictive control of the windshield wiper system.

[0030] Step S12: Based on the target vehicle's own vehicle operation data and the road topology information of the driving path, filter and process the preceding vehicle's operation data and preceding vehicle's environmental perception data to obtain the preceding vehicle's effective data set.

[0031] In this embodiment, the vehicle control system filters and processes the preceding vehicle's operating data and environmental perception data based on the target vehicle's own operating data and the road topology information of the driving path, thereby obtaining a valid set of preceding vehicle data. The self-operated vehicle data refers to the vehicle status parameters collected by the target vehicle itself in real time. These include, for example, the target vehicle's wiper frequency, rain sensor readings, vehicle speed, Global Navigation Satellite System (GNSS) position, and heading angle.

[0032] Road topology information refers to map or logical data that describes the spatial structure, lane distribution, and connectivity of the target vehicle's current location and the path ahead. Examples include lane line positions, number of lanes, lane connections, and traffic light positions in high-precision maps.

[0033] The effective data set of preceding vehicles refers to the subset of data on preceding vehicles that has been filtered and retained and is of reference value to the target vehicle. For example, it includes the wiper frequency and precipitation intensity data of preceding vehicles located in the same lane or adjacent lanes in the same direction.

[0034] It can be seen that the target vehicle uses its real-time location, speed, heading and other operational data, combined with the preset road space structure information, to perform logical judgment and filtering on all the received data of vehicles ahead, eliminating data located on invalid paths or with irrelevant relative positions, and retaining only the data of vehicles ahead located within the effective driving area of ​​the target vehicle.

[0035] For example, the target vehicle detects that ten connected vehicles ahead have sent wiper data. However, by comparing the vehicle's heading angle and road topology information, it finds that three of the vehicles are in the oncoming lane, two are on the right ramp of the fork in the road ahead, and the remaining five are in the vehicle's lane and the adjacent lane to its left. Therefore, the system removes the data of the first five invalid vehicles and stores only the data of the remaining five vehicles in the set of valid data of the vehicles ahead.

[0036] Thus, by precisely filtering the spatial dimension, interference data from irrelevant vehicles was eliminated, ensuring the accuracy and effectiveness of the data upon which subsequent environmental risk predictions were based.

[0037] Step S13: Based on the preceding vehicle's operating data, the vehicular operating data, and the set of valid preceding vehicle data, predict the environmental risk information of the road section ahead.

[0038] In this embodiment, the target vehicle predicts environmental risk information for the road ahead based on the preceding vehicle's operating data, its own operating data, and a set of valid preceding vehicle data. The environmental risk information for the road ahead refers to the assessment results of the severity of the environment or potential hazards faced by the target vehicle in the area it is expected to enter within the expected travel distance or in the expected future travel time.

[0039] It can be seen that the target vehicle integrates the valid data of the vehicle in front with its own operating status data, and performs calculations and analyses through specific algorithms or models to deduce the degree or possibility of the road environment ahead posing a threat to driving safety.

[0040] For example, the target vehicle obtains the wiper frequencies of three effective connected vehicles ahead, which are 40, 45, and 50 times per minute, respectively. Combined with the information that the target vehicle is currently traveling at a speed of 80 km / h and is 50 meters away from the vehicle in front, the built-in prediction model calculates that the precipitation risk coefficient of the target vehicle's location in the next 10 seconds is 0.75.

[0041] This enables a shift from static data reception to dynamic risk prediction, allowing vehicles to anticipate changes in rainfall conditions ahead.

[0042] Step S14: Based on the environmental risk information, generate joint control instructions for the target vehicle, wherein the joint control instructions are used to drive and control multiple actuators of the target vehicle.

[0043] In this embodiment, the target vehicle generates a joint control command based on environmental risk information. This joint control command is used to drive and control multiple actuators of the target vehicle. The joint control command refers to an integrated operation command generated by the target vehicle based on the risk information, containing multiple control sub-instructions. The multiple actuators refer to hardware devices on the target vehicle capable of responding to commands and changing their operating state, such as wiper motors, headlight relays, or air conditioning compressors.

[0044] It can be seen that the target vehicle maps and combines one or more control signals according to the environmental risk level or value and according to the preset logical rules. The signal is intended to simultaneously command different hardware components inside the vehicle to work together to cope with the current or upcoming environmental conditions.

[0045] For example, when environmental risk information is determined to be high risk, the target vehicle generates a joint control command. This command is simultaneously sent to the wiper controller to start high-speed wiping, to the lighting controller to turn on the fog lights, and to the air conditioning controller to turn on the defogging mode via the vehicle network, thereby achieving synchronous response of multiple systems.

[0046] Thus, through the coordinated operation of multiple actuators, the response speed and overall safety of visibility and vehicle visibility are improved when driving in rainy weather.

[0047] In summary, this application provides a vehicle control method applied to a target vehicle. The method includes: first, receiving preceding vehicle operation data and preceding vehicle environmental perception data, wherein the preceding vehicle includes at least one connected vehicle located ahead of the target vehicle's driving path. This allows the target vehicle to acquire beyond-line-of-sight real-world road information, providing foundational data support for subsequent predictive control; second, filtering the preceding vehicle operation data and preceding vehicle environmental perception data based on the target vehicle's own vehicle operation data and the road topology information of the driving path to obtain a set of valid preceding vehicle data. Thus, through precise spatial filtering, invalid data is excluded. Interference data from vehicles is used to ensure the accuracy and effectiveness of the data used for subsequent environmental risk prediction. Then, based on the operating data of the preceding vehicle, the vehicle's own operating data, and the effective data set of the preceding vehicle, environmental risk information of the road section ahead is predicted. This realizes the transformation from static data reception to dynamic risk prediction, enabling vehicles to know the trend of environmental changes ahead in advance. Finally, based on the environmental risk information, joint control commands for the target vehicle are generated. These joint control commands are used to drive and control multiple actuators of the target vehicle. Thus, through the coordinated linkage of multiple actuators, the response speed of driving visibility and vehicle visibility is improved, as well as overall safety.

[0048] This application achieves the goal of driving and controlling the actuators of the vehicle based on the network communication between vehicles, combining the operating data of the preceding vehicle, the environmental perception data of the preceding vehicle, the operating data of the vehicle itself, and the road topology information, so as to avoid environmental risks on the road ahead in advance. This achieves the technical effect of enhancing the real-time prediction of environmental risks on the road ahead, thereby improving driving safety and comfort. It also solves the technical problem of the existing technology, which suffers from delayed environmental risk perception and poor driving safety and comfort due to the limited perception range of a single vehicle.

[0049] Optionally, the preceding vehicle's operating data and the preceding vehicle's environmental perception data are obtained by parsing the wiper status messages broadcast by the preceding vehicle at a target frequency, wherein the target frequency is dynamically adjusted by the preceding vehicle according to the wiper's working mode. In this embodiment of the application, the wiper status message refers to a digital data message containing key information such as the wiper operating frequency and rainfall intensity of the vehicle ahead.

[0050] Target frequency refers to the number of times a vehicle ahead broadcasts the message to its surroundings per unit time. Wiper operating mode refers to the current operating status classification of the wiper system, such as wipers off, low-speed intermittent wiping, or high-speed continuous wiping.

[0051] It can be seen that the vehicle in front adaptively changes the data broadcast transmission rate according to the specific working mode of its windshield wipers, and sends status messages containing operational and environmental perception data at this rate. The target vehicle obtains the required data by receiving and parsing these messages at different frequencies.

[0052] For example, the broadcast strategy is intelligent and dynamic. When the windshield wipers of the vehicle in front are off, a message containing zero-value wiper data is broadcast every 5 seconds. When rain suddenly occurs and the wipers switch to high-speed mode, the vehicle in front immediately increases the broadcast frequency to once per second and sends a message containing high wiper frequency and heavy rainfall intensity data. The target vehicle receives and parses the message to obtain the latest data.

[0053] Thus, by dynamically adjusting the broadcast frequency, the optimal allocation of communication resources under normal and abnormal operating conditions was achieved, and the timely updating of key environmental data was ensured.

[0054] Optionally, the wiper status message also includes a message identifier field and a data confidence field.

[0055] In this embodiment, the message identifier field refers to identification information such as a sequence number or timestamp used to uniquely distinguish or track a specific data packet. Examples include an incrementing message sequence number, a hash value generated based on a timestamp, or a unique vehicle identifier. The data confidence field refers to a numerical indicator used to quantify the reliability, accuracy, or freshness of the currently collected data. Examples include floating-point numbers between 0 and 1 (e.g., 0.95 indicates high confidence) or grading labels (e.g., high, medium, low). Its calculation may be based on sensor signal-to-noise ratio, data latency, or sensor self-test status.

[0056] It can be seen that when the vehicle control system generates wiper status messages, in addition to the core operating and environmental data, it also adds identifiers to identify the uniqueness or order of the messages and a reliability value to evaluate the data quality, so that the receiving end can verify the data during processing.

[0057] For example, the vehicle's local vehicle control system continuously fuses raw data into a single "wiper status message." This message's data structure is carefully designed, containing the following key fields: message sequence number (for replay protection), vehicle unique identifier (anonymized), high-precision GNSS position and heading angle, vehicle instantaneous speed, current wiper operating mode (intermittent / low-speed continuous / high-speed continuous) and the actual wiping frequency value after Kalman filtering smoothing, an estimated precipitation intensity value output from the local rain sensor (unit: mm / hour), and a "data confidence" factor (range 0-1) calculated by combining sensor signal-to-noise ratio and data freshness. This factor determines whether the wiper frequency data in the message has high reference value and can be included in subsequent fusion calculations.

[0058] Therefore, by providing unique message identifiers and data quality assessment indicators, the receiver's ability to judge the validity and reliability of data is enhanced, thereby improving the robustness of the vehicle control system.

[0059] Optionally, the method further includes: Step S15: In response to receiving an emergency event message broadcast by the vehicle ahead, the emergency event message is parsed to obtain updated vehicle ahead operation data and updated vehicle ahead environmental perception data. The emergency event message is generated by the vehicle ahead when it detects that the rate of change of the windshield wiper operating frequency within a preset time period is greater than a preset threshold. The priority of the emergency event message is higher than that of the windshield wiper status message.

[0060] In this embodiment, when an emergency event message broadcast by a vehicle ahead is received, the target vehicle parses the emergency event message to obtain updated preceding vehicle operation data and updated preceding vehicle environmental perception data. The emergency event message is generated by the preceding vehicle when it detects that the rate of change of the windshield wiper operating frequency within a preset time period exceeds a preset threshold. The emergency event message has a higher priority than the windshield wiper status message. An emergency event message can be understood as a specific data message with high priority generated by the preceding vehicle when it detects a drastic change in the windshield wiper operating frequency within a short period. For example, it may include an Ethernet message containing high-frequency wiper data, heavy rainfall intensity, and a high confidence factor.

[0061] The preset duration refers to a fixed time window used to calculate the rate of change of wiper frequency. The preset threshold refers to the critical value for judging whether the change of wiper frequency constitutes an emergency event. For example, the preset duration is 3 seconds or other custom durations, and the preset threshold is the wiper frequency jumping from 20 times / minute to 45 times / minute or other custom frequency changes within 3 seconds.

[0062] Priority refers to the hierarchical standard that determines the order of message transmission and processing in a communication system. Higher priority means that messages are processed and transmitted first.

[0063] It can be seen that after the target vehicle receives a high-priority emergency event message sent by the vehicle in front, it decodes and processes it to obtain the latest operating status of the vehicle in front and environmental perception data. The trigger condition for this message is that the wiper frequency of the vehicle in front changes beyond the set limit within a specified time, and its processing order in the system takes precedence over the regular wiper status message.

[0064] For example, when the vehicle control system detects that the windshield wiper operating frequency rises sharply within a short period of time (e.g., within 3 seconds) above its threshold (e.g., from 20 times / minute to 45 times / minute), or when the local rain sensor reading increases sharply, an emergency event broadcast is triggered. This broadcast message has the highest priority and immediately increases the broadcast frequency to 5Hz for a period of time to ensure that critical warning information is effectively captured.

[0065] Thus, the high-priority emergency messaging mechanism ensures that vehicles behind can obtain critical early warning data immediately in the event of sudden severe weather, reducing information delays.

[0066] Optionally, the vehicles ahead include multiple target vehicles; based on the vehicle's own operating data and road topology information, the operating data and environmental perception data of the vehicles ahead are filtered and processed to obtain the effective data set of the vehicles ahead, including: Step S121: Extract the target vehicle's current position, speed, and heading angle from the vehicle's operating data.

[0067] Step S122: Traverse multiple object vehicles. For the current object vehicle in the traversal, extract the vehicle position, vehicle speed, vehicle heading angle and message timestamp of the current object vehicle at the time of message generation from the running data of the preceding vehicle corresponding to the current object vehicle.

[0068] Step S123: Calculate the expected position of the current object vehicle at the current moment based on the vehicle position, vehicle speed, vehicle heading angle and message timestamp.

[0069] Step S124: Based on the road topology information, vehicle position, vehicle speed, and vehicle heading angle, determine whether the expected position is within the valid area corresponding to the driving path.

[0070] Step S125: If the expected location is within the valid area, add the preceding vehicle's operating data and the preceding vehicle's environmental perception data corresponding to the current target vehicle to the preceding vehicle's valid data set.

[0071] The target vehicle can be any connected vehicle in front that is within the communication range of the receiving vehicle via a V2V communication link. In this embodiment, the target vehicle specifically refers to those vehicles in front that broadcast wiper status messages or emergency event messages. The vehicle control system needs to process and filter the data of these vehicles one by one in order to extract valid data of the vehicles in front.

[0072] In this embodiment, the target vehicle first extracts its current position, speed, and heading angle from its vehicle operation data. Here, the vehicle position refers to the target vehicle's current geographic coordinates in the Earth coordinate system, the vehicle speed refers to the target vehicle's current travel speed, and the heading angle refers to the angle between the target vehicle's current travel direction and true north.

[0073] It can be seen that the target vehicle reads and separates three key parameters from its own real-time operational data: its current geographical location, driving speed, and driving direction, which serve as the basis for subsequent spatial calculations. For example, the target vehicle's cockpit domain controller reads data from the GNSS module and vehicle speed sensor to obtain that the target vehicle is currently located at 39.90 degrees north latitude and 116.40 degrees east longitude, traveling at a speed of 60 km / h, with its front facing northeast (heading angle of 45 degrees).

[0074] Therefore, by providing a precise spatial baseline for the vehicle, it provides the necessary reference coordinates for subsequently determining the relative positional relationship between the vehicle and other vehicles.

[0075] Secondly, the target vehicle iterates through multiple target vehicles. For the current target vehicle during the iteration, the vehicle position, vehicle speed, vehicle heading angle, and message timestamp at the time of message generation are extracted from the preceding vehicle's running data.

[0076] The message generation time refers to the point in time when the preceding vehicle sends the windshield wiper status message, and the message timestamp is a time stamp embedded in the message that records the message generation time. For example, if the current object vehicle is the preceding vehicle with ID "Car_005", the message generation time is 10:00:00.100 UTC (meaning that this time uses Coordinated Universal Time, not the local time of the vehicle's location (such as Beijing time UTC+8)), the vehicle location is the latitude and longitude of the preceding vehicle at this moment, and the message timestamp is 10:00:00.100. This data exists in the received V2V message.

[0077] It can be seen that the target vehicle processes the data received from each preceding vehicle in sequence. For each vehicle that is traversed, it extracts the vehicle's geographical location, speed, direction of travel, and the specific timestamp of the message at the moment the message was sent from its corresponding data packet.

[0078] This enabled the orderly extraction of historical status data for each preceding vehicle, providing raw data support for subsequent location estimation based on time differences.

[0079] Then, the target vehicle calculates the expected position of the current target vehicle at the current moment based on the vehicle position, vehicle speed, vehicle heading angle and message timestamp.

[0080] The expected position refers to the theoretical position of the vehicle at the time of receipt (i.e., the current time) based on the state of the preceding vehicle at the time of message generation and the time difference, assuming that it will maintain its original motion state. The vehicle position, vehicle speed, vehicle heading angle and message timestamp are used to construct a kinematic model for position estimation.

[0081] It can be seen that the target vehicle uses the position, speed, and heading of the preceding vehicle when the message was sent, as well as the time difference between the message timestamp and the time the target vehicle receives the message, to estimate the theoretical geographical location of the preceding vehicle at the current moment through kinematic calculations.

[0082] For example, the message timestamp is 10:00:00.100, the current time is 10:00:00.200, the time difference is 100ms, the speed of the vehicle in front is 70km / h, the heading is 45 degrees, the vehicle control system calculates that the vehicle in front has moved forward by about 1.94 meters in these 100ms, thus obtaining the expected position of the vehicle in front at the current moment as 39.9101 degrees north latitude and 116.4101 degrees east longitude.

[0083] This eliminates the location error caused by message transmission delay, enabling the receiver to accurately assess the actual spatial position of the vehicle ahead based on a unified time reference.

[0084] Next, the target vehicle determines whether its expected location is within the valid area corresponding to the driving path based on road topology information, its own position, its own speed, and its own heading angle.

[0085] The effective area refers to the spatial range on the driving path that has reference value for the target vehicle, such as the area of ​​the current lane and the left lane within 200 meters ahead. The area after the opposite lane or the ramp is usually excluded from the effective area.

[0086] As can be seen, the target vehicle spatially compares the calculated expected position of the vehicle ahead with the effective driving path area determined based on its own vehicle status and road map data to determine whether the vehicle ahead is within a spatial range meaningful to its driving decisions. For example, based on its own heading of 45 degrees and a high-precision map, the target vehicle determines its current driving path to be its own lane and the lane to its left. Checking the expected position of the vehicle ahead, it finds that the vehicle is 50 meters ahead of its own lane and determines that it is within the effective area. If another vehicle is in the oncoming lane, it is determined that the vehicle is not within the effective area.

[0087] Thus, by using a spatial filtering mechanism, data interference from irrelevant vehicles was eliminated, ensuring the relevance and accuracy of subsequent data processing.

[0088] Finally, if the expected location is within the valid area, the target vehicle will add the preceding vehicle's operating data and the preceding vehicle's environmental perception data to the preceding vehicle's valid data set.

[0089] As can be seen, when the expected position of a vehicle in front is confirmed to be within the effective driving path area of ​​the target vehicle, the target vehicle formally includes the corresponding operating status data and environmental perception data into the effective data set for use by subsequent algorithms. For example, if the vehicle control system confirms that the vehicle with ID "Car_005" is within the effective path in front of the vehicle, it will store data such as the vehicle's wiper frequency of 45 times / minute and rainfall intensity of 50mm / h into the "effective data set of the vehicle in front", while ignoring the data of other vehicles in the oncoming lane.

[0090] This completed the final step of data filtering, constructing a high-quality input dataset and laying the data foundation for accurate prediction of environmental risks on the road ahead.

[0091] Optionally, environmental risk information includes precipitation risk coefficients; based on preceding vehicle operating data, voluntary vehicle operating data, and the effective set of preceding vehicle data, predicted environmental risk information includes: Step S131: Based on the valid data set of the preceding vehicles, determine the wiper operating frequency sequence and precipitation intensity estimation sequence corresponding to the valid target vehicles.

[0092] Step S132: Based on the preceding vehicle's operating data and the vehicle's operating data, determine the relative distance sequence corresponding to the valid target vehicles.

[0093] Step S133: Normalize the wiper operating frequency sequence, precipitation intensity estimation sequence, relative distance sequence, and the target vehicle's own environmental parameters to obtain a multidimensional feature vector.

[0094] Step S134: Input the multidimensional feature vector into the target prediction model for inference to obtain the precipitation risk coefficient. The target prediction model is a deep neural network model trained using multiple sets of sample data.

[0095] In this embodiment of the application, precipitation risk is used to characterize the likelihood and intensity level of the target vehicle encountering the severity of rainfall or the deterioration of rain conditions on the road ahead within the expected travel time (such as the next 10-30 seconds). This coefficient usually takes a value between 0 and 1, and the higher the value, the worse the rain conditions ahead or the greater the risk.

[0096] In this embodiment, the target vehicle first determines the windshield wiper operating frequency sequence and precipitation intensity estimation sequence corresponding to the valid target vehicle based on the valid preceding vehicle data set. The valid target vehicle refers to the preceding vehicle located on the valid driving path of the target vehicle after screening.

[0097] The wiper operating frequency sequence refers to a list of data on the frequency of the wipers over a period of time, such as an array of frequency values ​​recorded per second over the past 10 seconds [20, 25, 30, 40, 50, ...]. The precipitation intensity estimation sequence refers to a list of data on the amount of rainfall estimated by the vehicle through sensors over a corresponding time period, such as an array of rainfall intensity values ​​at a corresponding time point [2, 3, 5, 8, 10, ...] mm / h.

[0098] It can be seen that the target vehicle extracts the wiper frequency change data and precipitation intensity change data of each valid preceding vehicle from the selected valid preceding vehicle data within the historical time window, forming time series data.

[0099] Therefore, time series data reflecting the changing trend of rainfall conditions ahead were constructed, providing a basic input for the model to understand the dynamic evolution of weather.

[0100] Then, based on the preceding vehicle's operating data and its own vehicle's operating data, the target vehicle determines the relative distance sequence corresponding to the valid target vehicles. The relative distance sequence refers to the data series showing the distance changes between the target vehicle and each valid target vehicle at different times. The relative distance can be a numerical value, such as 50 meters, 100 meters, or 150 meters. For example, a relative distance sequence could be an array of distances between the target vehicle and "Car_001" over the past 10 seconds [150, 140, 130, 120, 110, 100, 90, 80, 70, 60], indicating that the two vehicles are gradually approaching each other.

[0101] It can be seen that the vehicle control system uses the position data of the target vehicle and the effective preceding vehicle at historical moments to calculate the spatial distance between the two vehicles at each moment and form a distance sequence that changes over time.

[0102] This quantifies the spatial proximity between the vehicle in front and the vehicle itself, helping the model assess the impact of the rain conditions of the vehicle in front on the vehicle itself.

[0103] Next, the target vehicle normalizes the wiper operating frequency sequence, precipitation intensity estimation sequence, relative distance sequence, and the target vehicle's own environmental parameters to obtain a multi-dimensional feature vector.

[0104] Normalization refers to the data preprocessing process that maps values ​​of different magnitudes and units to the same standard range (such as between 0 and 1 or between -1 and 1). For example, it maps the wiper frequency of 20-60 times / minute to 0.0-1.0, and the distance of 60-150 meters to 1.0-0.0. An example of vehicle environmental parameters is the current speed of the vehicle at 80 km / h and the temperature at 25℃.

[0105] A multidimensional feature vector is a high-dimensional data array composed of processed wiper frequency, precipitation intensity, relative distance, and vehicle environmental parameters. Vehicle environmental parameters include vehicle speed, ambient temperature, and light intensity.

[0106] It can be seen that the vehicle control system extracts the wiper frequency sequence, precipitation intensity sequence, relative distance sequence, and the vehicle's own environmental parameters, performs mathematical scaling to make their numerical range consistent, and combines them into a feature vector of a unified format for input into the neural network.

[0107] For example, the vehicle control system normalizes the wiper frequency sequence to obtain array A, the precipitation intensity sequence to obtain array B, the relative distance sequence to obtain array C, and the vehicle parameters to obtain scalar d. Finally, all arrays and scalars are concatenated into a long vector V=[A, B, C, d].

[0108] This eliminates the impact of differences in data dimensions on model training and inference, and improves the stability and convergence speed of deep neural network processing.

[0109] Finally, the target vehicle inputs the multi-dimensional feature vector into the target prediction model for inference to obtain the precipitation risk coefficient. The target prediction model is a deep neural network model trained using multiple sets of sample data.

[0110] As can be seen, by inputting the preprocessed multidimensional feature vector into the trained deep neural network model, the model outputs a coefficient value that represents the magnitude of precipitation risk on the road ahead through its internal calculation logic.

[0111] For example, for data from multiple vehicles ahead in the selected direction of travel, the vehicle control system can employ a fusion prediction model based on a deep neural network (DNN). The model's input consists of spatiotemporally aligned and normalized sequences of windshield wiper frequencies from preceding vehicles, rainfall data sequences, relative distances, and the vehicle's environmental parameters. The model output is a scalar value (0-1) called the "Precipitation Risk Coefficient for Ahead Road Section," which comprehensively characterizes the severity of rainfall the vehicle will encounter within a predicted travel time (e.g., 10-30 seconds). During training, the model utilizes extensive paired data collected from real vehicles, including the behavior of preceding vehicle wipers and the actual rainfall conditions encountered by the vehicle, thereby learning the complex mapping relationship between preceding vehicle wiper dynamics and real-world weather propagation.

[0112] Therefore, by using the quantitative model output, an intuitive and comparable decision-making basis is provided for subsequent graded early warning and proactive control strategies.

[0113] Optionally, based on environmental risk information, the generation of joint control instructions includes: Step S141: Determine the warning level based on environmental risk information.

[0114] Step S142: Based on the warning control strategy corresponding to the warning level, generate multiple sub-control commands corresponding to multiple actuators, including the wiper system, the lighting system, and the air conditioning system.

[0115] In this embodiment, the vehicle control system first determines the warning level based on environmental risk information. The warning level refers to different safety alert levels divided according to the magnitude of the risk coefficient, typically used to distinguish between low, medium, and high levels of risk in order to take corresponding countermeasures. For example, it includes Level 1 warning (low risk, such as coefficient 0.3-0.6), Level 2 warning (medium risk, such as coefficient 0.6-0.8), and Level 3 warning (high risk, such as coefficient > 0.8).

[0116] As can be seen, the vehicle control system compares the calculated environmental risk information (such as the precipitation risk coefficient) with the preset risk threshold range to determine which warning level it belongs to.

[0117] This allows for the classification and grading of risks, providing a clear logical basis for matching specific control strategies in the future.

[0118] Then, the vehicle control system will generate multiple sub-control commands corresponding to multiple actuators based on the warning control strategy corresponding to the warning level. These actuators include the wiper system, the lighting system, and the air conditioning system.

[0119] Among them, sub-control commands refer to control signals or commands sent to various specific hardware actuators. Examples of sub-control commands include wiper frequency control messages sent to the Body Control Module (BCM), headlight switch signals sent to the BCM, and airflow adjustment signals sent to the Heating, Ventilation, and Air Conditioning (HVAC) controller. The actuators include wiper motors, headlight controllers, air conditioning compressors, and damper motors.

[0120] It can be seen that the vehicle control system searches for the corresponding preset control strategy based on the currently determined warning level, and decomposes it into specific control commands for the wiper system, lighting system and air conditioning system.

[0121] This enables the automated generation of multi-system collaborative control, ensuring that vehicle components can respond in unison according to predetermined safety strategies under different risk levels.

[0122] Optionally, the early warning and control strategy includes: When the warning level is low risk, the first sub-control command is generated. The first sub-control command is used to control the wiper system to perform the preparation mode switching action and the wiper blade reset action.

[0123] When the warning level is medium risk, a second sub-control command is generated. This second sub-control command is used to control the wiper system to perform wiping actions at a first wiping frequency determined based on the environmental risk information.

[0124] When the warning level is high-risk, a third sub-control command, a fourth sub-control command, and a fifth sub-control command are generated. The third sub-control command is used to control the wiper system to switch to the target continuous wiping mode. The second wiping frequency corresponding to the target continuous wiping mode is higher than the first wiping frequency. The fourth sub-control command is used to control the lighting system to turn on the front and rear fog lights. The fifth sub-control command is used to control the air conditioning system to switch to the windshield defroster mode.

[0125] In this embodiment of the application, the low risk level refers to the state in which environmental risk information (such as precipitation risk coefficient) is within a preset low threshold range. At this time, the preparation mode switching action refers to the logical state change of the wiper control system from passive sensing mode to active predictive control mode. The wiper blade reset action refers to the operation of driving the wiper arm to the starting position at the bottom of the windshield.

[0126] It can be seen that when the risk level is determined to be low, the vehicle control system generates the first sub-control command, which includes two actions: first, switching the working logic mode of the wipers, and second, physically resetting the wiper blades to the standby position.

[0127] For example, at a low-risk level, i.e., Level 1 warning (risk coefficient of 0.3-0.6), the vehicle control system uses the instrument cluster or head-up display (HUD) to indicate to the driver the possibility of rain ahead with non-intrusive icons (such as gradient cloud and rain icons). At the same time, the automatic wiper system switches from a "purely reactive" mode to a "predictive preparation" mode, pre-charges the wiper motor drive circuit, and pre-positions the wiper blades to the optimal activation position at the lower edge of the windshield, reducing mechanical response delay to near zero.

[0128] Thus, in low-risk situations, the wiper system can be preheated and its position preset, shortening the mechanical response time during subsequent sudden rainfall.

[0129] The medium-risk level refers to a state where environmental risk information is within a preset threshold range. The second sub-control command is a specific control command generated by the generation module and sent to the wiper system when the vehicle control system determines that the warning level is medium-risk. This command instructs the wiper system to perform active wiping operations according to a first wiping frequency calculated based on the environmental risk information. The first wiping frequency refers to the wiper operating speed calculated based on the current environmental risk information and applicable to moderate rain conditions. The wiping action refers to the wiper motor driving the wiper blades to reciprocate on the windshield. For example, the first wiping frequency is 30 times / minute.

[0130] It can be seen that when the risk level is determined to be medium, the vehicle control system generates a second sub-control command, which instructs the wiper system to actively wipe at a specific frequency (first wiping frequency) calculated based on environmental risk information.

[0131] For example, at a medium-risk level, i.e., a level 2 warning (risk coefficient 0.6-0.8), the system issues a clear visual and level 1 audio alert (such as a single "beep" sound). Simultaneously, the wiper controller instructs the wiper motor to operate at a predicted frequency calculated based on the risk coefficient and a weighted average of data from the vehicle ahead. For instance, the vehicle control system might preemptively set the wipers to a low-to-medium speed continuous wiping mode even before the windshield is wet with raindrops.

[0132] Therefore, at medium-risk levels, the wiper speed can be predictively adjusted to maintain clear visibility and avoid blurred vision before the rain intensifies.

[0133] A high-risk level refers to a state where environmental risk information exceeds a preset high threshold. The third sub-control command is a specific control command generated by the generation module and sent to the wiper system when the system determines the warning level to be high-risk. This command instructs the wiper system to switch to the target continuous wiping mode and perform high-intensity wiping at a second wiping frequency higher than that of the medium-risk level. The target continuous wiping mode refers to the working mode in which the wipers continuously operate at the highest or higher speed, and the second wiping frequency refers to the highest wiping speed corresponding to this mode.

[0134] The fourth sub-control command refers to the specific control command generated by the generation module and sent to the lighting system when the system determines that the warning level is high-risk. This command enables the indicator light system to turn on the vehicle's front fog lights and rear fog lights to improve the vehicle's visibility and lighting capabilities in low-visibility environments.

[0135] The fifth sub-control command refers to the specific control command generated by the generation module and sent to the air conditioning system when the system determines that the warning level is high-risk. This command instructs the air conditioning system to switch to the windshield defrost mode, which quickly eliminates fog or condensation on the inside of the windshield by adjusting the airflow direction and volume.

[0136] It can be seen that when the risk level is determined to be high, the vehicle control system generates three sub-control commands at the same time: the third command controls the wipers to wipe continuously at a high frequency, the fourth command controls the lighting system to turn on the fog lights, and the fifth command controls the air conditioning system to defog.

[0137] For example, at a high-risk level, i.e., a Level 3 emergency warning (risk coefficient > 0.8), the vehicle control system issues a strong visual and auditory warning (such as continuous audible alerts combined with flashing icons). The windshield wipers automatically switch to the highest speed continuous wiping mode. Simultaneously, the vehicle control system can coordinate with other vehicle body controllers to automatically activate the front and rear fog lights, switch the air conditioning to windshield defroster mode, and upload the "heavy rain section" information marked by the vehicle via the vehicle-to-everything (V2X) cloud platform for reference by traffic management systems in a wider area.

[0138] It can be seen that when the risk level is determined to be high, the multi-system linkage of windshield wipers, lights, and air conditioning ensures the best possible driving visibility and vehicle visibility under extreme rain conditions.

[0139] Optionally, the method further includes: Step S15: After the target vehicle enters the road section ahead, determine the actual rainfall intensity and record the corresponding intervention and control information for the target vehicle.

[0140] Step S16: Determine the model prediction bias based on the actual precipitation intensity and precipitation risk coefficient.

[0141] Step S17: Construct feedback sample data based on intervention control information and model prediction bias.

[0142] Step S18: Use the feedback sample data to fine-tune the parameters of the target prediction model in order to update the target prediction model.

[0143] In this embodiment, firstly, after the target vehicle enters the road section ahead, the vehicle control system determines the actual rainfall intensity and records the corresponding intervention control information for the target vehicle. The actual rainfall intensity refers to the real rainfall data detected at the current moment by the rain sensor mounted on the target vehicle. The intervention control information refers to any manual modification, cancellation, or confirmation of sub-control commands output by the driver during the automatic warning control process executed by the vehicle control system, such as manually turning off the windshield wipers or adjusting the wiper frequency.

[0144] As can be seen, when the vehicle actually enters the rainfall area predicted by the data from the vehicle ahead, the vehicle control system uses local sensors to obtain the actual rainfall intensity data, and at the same time collects data on the driver's intervention in the automatic control behavior of the vehicle control system. For example, when the vehicle enters the predicted rainfall area, the local rain sensor detects that the actual rainfall intensity is 40 mm / h. At the same time, the system records that after receiving the level 2 warning, the driver manually turned down the wiper frequency, and this operation is recorded as intervention control information.

[0145] This yielded data on the discrepancy between the vehicle control system's predictions and the actual situation, as well as user preference data, providing fundamental information for model optimization.

[0146] Secondly, the vehicle control system determines the model prediction bias based on the actual precipitation intensity and precipitation risk coefficient. The model prediction bias refers to the difference or error between the predicted rainfall severity represented by the precipitation risk coefficient output by the deep neural network and the actual rainfall severity represented by the actual precipitation intensity obtained from local rain gauges.

[0147] As can be seen, the vehicle control system compares the actual precipitation intensity with the previously calculated precipitation risk coefficient to quantify the accuracy of the model's prediction and determine the prediction bias. For example, if the vehicle control system previously calculated a precipitation risk coefficient of 0.6 (corresponding to medium risk), while the current actual precipitation intensity is only 10 mm / h (low risk), the vehicle control system calculates a large positive deviation between the two, indicating that the model overestimated the rainfall intensity.

[0148] This quantifies the error of the prediction model, providing a specific basis for subsequent adjustments to the model parameters.

[0149] Then, the vehicle control system constructs feedback sample data based on intervention control information and model prediction bias. This feedback sample data refers to training data units formed by combining model prediction bias with driver intervention control information. This data contains three layers of information: "model prediction results," "real-world results," and "driver correction results," which are used to guide the model's self-learning.

[0150] As can be seen, the vehicle control system associates and packages intervention control information with model prediction bias to form a complete feedback sample data record. For example, the vehicle control system packages "prediction bias: high" with "intervention operation: manual frequency reduction" to form a feedback sample data, which is stored in the local log database for later use in model training.

[0151] Thus, a complete training sample containing model error and user preferences was constructed, realizing closed-loop data accumulation from prediction to execution to feedback.

[0152] Finally, the vehicle control system uses the feedback sample data to fine-tune the parameters of the target prediction model in order to update the target prediction model.

[0153] Among them, parameter fine-tuning refers to making small adjustments to the internal parameters of the deep neural network, such as weights and biases, using newly constructed feedback sample data and algorithms such as backpropagation.

[0154] It can be seen that the vehicle control system uses feedback sample data to retrain or update the parameters of the target prediction model, enabling the model to self-optimize based on past prediction errors and user intervention records. For example, the system reads feedback sample data from the past week, fine-tunes the DNN model, corrects the weights that previously overestimated the rainfall intensity, and thus generates an updated model version.

[0155] For example, the vehicle control system incorporates a lightweight learning module. This module records the deviation between the actual rain conditions encountered by the vehicle (verified by local sensors) and the predicted risk coefficient after each warning is triggered, as well as any manual intervention by the driver in the vehicle control system's automatic adjustment behavior (such as the driver forcibly lowering the wiper speed automatically set by the vehicle control system). This feedback data is used to fine-tune the parameters of the local DNN prediction model (using a federated learning approach), enabling the vehicle control system to gradually adapt to common local weather patterns, the wiper characteristics of mainstream vehicles in the fleet, and the driver's personal preferences, achieving increasingly accurate and personalized optimization over time.

[0156] This enables the model to evolve adaptively, making the predictions more consistent with actual road conditions and user habits, and improving the accuracy of warnings.

[0157] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, use and processing of the relevant data must comply with the relevant laws, regulations and standards of the relevant countries and regions, and corresponding operation entry points are provided for users to choose to authorize or refuse.

[0158] According to an embodiment of this application, a vehicle control system embodiment is provided. Figure 2 This is a structural block diagram of a vehicle control system according to an embodiment of this application.

[0159] It should be noted that this system is applied to the target vehicle, operating as an embedded network node in each vehicle. Its hardware consists of an enhanced data acquisition unit, a dedicated V2V communication module, a domain controller-level data processing unit, and intelligent actuators, enabling the execution of the aforementioned vehicle control methods. For example... Figure 2 As shown, the system includes: The enhanced data acquisition unit 201 is a deeply integrated wiper motor controller. This controller not only outputs control signals but also functions as a sensor, using high-precision Hall effect sensors to monitor the precise angular position and instantaneous rotation speed of the wiper arm in real time. It calculates the current actual operating frequency of the adaptive wipers (in times per minute) at a sampling rate of at least 20Hz. Simultaneously, it acquires vehicle speed, ambient temperature, and light intensity signals from the vehicle bus and obtains high-precision timestamps and latitude / longitude coordinates (positioning accuracy better than 1.5 meters) via a Global Navigation Satellite System receiver.

[0160] The dedicated V2V communication module 202 employs hardware units that support direct communication via Cellular Vehicle-to-Everything (C-V2X PC5) or Dedicated Short-Range Communications (DSRC) protocols, operating in the 5.9GHz intelligent transportation system frequency band. Its transmit power can be dynamically adjusted between 100 milliwatts and 1 watt to adapt to different communication distance requirements.

[0161] The data processing unit 203, which is concurrently the vehicle cockpit domain controller or the autonomous driving domain controller, is equipped with multi-core processing and runs a lightweight real-time operating system (RTOS) or Linux system, and is responsible for running the core algorithm of this invention.

[0162] The intelligent actuator 204, which is the vehicle's windshield wiper system assembly, requires its motor drive to support direct frequency and gear control based on controller local area, message or Ethernet messages, with a response latency of less than 50 milliseconds.

[0163] Figure 3 This is a flowchart illustrating another vehicle control method according to an embodiment of this application, using a pure electric SUV equipped with a high-performance domain controller and a C-V2X communication module as an example.

[0164] like Figure 3 As shown, the steps of this application embodiment include: step S31, hardware deployment and initialization; step S32, data acquisition and message generation; step S33, intelligent broadcasting and receiving processing; step S34, fusion prediction and active control execution; and step S35, system testing and verification.

[0165] Specifically, the hardware deployment and initialization include: firstly, integrating a high-sensitivity optical rain sensor into the base of the rearview mirror above the vehicle's windshield.

[0166] Then, the original windshield wiper motor controller is upgraded to use an intelligent motor drive chip that supports direct angle and speed feedback. The real-time wiping frequency data (e.g., with an accuracy of 0.1 times / minute) is sent to the cabin domain controller through a dedicated Local Interconnect Network (LIN) bus or high-speed bus channel.

[0167] Next, the cockpit domain controller uses a processor with a computing power of no less than 20K DMIPS, on which runs an automotive open system architecture adaptive platform or a customized operating system. The controller connects to the V2X communication module via an Ethernet interface and interacts with the vehicle network and powertrain network via a bus.

[0168] Finally, after the vehicle control system is powered on, it first completes a self-test, including checking the cleanliness of the rain sensor mirror, the feedback path of the wiper motor, the quality of the GNSS signal, and the status of the V2X communication link.

[0169] Step S32, data acquisition and message generation includes: First, a dedicated service within the domain controller reads the feedback message from the wiper motor controller at a frequency of 20Hz and parses the instantaneous frequency. Simultaneously, the analog output of the rain gauge was read at a frequency of 10Hz and converted into precipitation intensity using a lookup table method. .

[0170] Then, every 100 milliseconds (0.1 seconds), the vehicle control system performs a data packetization. It obtains the latest latitude and longitude (lat, on), speed, and heading from the GNSS receiver, along with a UTC timestamp accurate to milliseconds. The core algorithm module then bases this on the 10 data points from the past second. The sampled values ​​are filtered using a first-order hysteresis filter to calculate a smooth, jitter-resistant reporting frequency. . and The vehicle status and other information are entered into a message template in a preset ASN.1 encoding format, and a confidence level is calculated. The confidence level takes into account factors such as whether the rain sensor is obstructed by dirt (determined by the reflection intensity of the built-in infrared light-emitting diode, LED) and whether the wiper blades are worn out, causing abnormal wiping resistance (determined by motor current feedback).

[0171] Step S33, intelligent broadcasting and receiving processing includes: First, the vehicle control system sends the encapsulated message to the V2X protocol stack and broadcasts it according to a preset strategy. On a clear, rainless day, the vehicle control system sends a heartbeat message every 5 seconds, with both the wiper frequency and rainfall fields in the message being zero.

[0172] Then, when the vehicle's windshield wipers activate, the vehicle control system immediately switches to the 1Hz regular broadcast mode. If the algorithm detects... If the increase exceeds 50% within two consecutive cycles, an event message will be generated and broadcast immediately, and the broadcast frequency will be temporarily increased to 5Hz for 3 seconds.

[0173] Finally, on the receiving side, another service within the cockpit domain controller of the vehicle control system listens for V2X air interface messages. For each received message, it first performs digital signature verification to ensure the source is trustworthy. Then, it calls the "spatiotemporal alignment" function library: assuming a communication delay of " The message was sent from a location where the sender was located at the time. The speed is The heading is This vehicle is currently located at The speed is The heading is The algorithm uses the World Geodetic System 1984 (WGS84) to extrapolate the sender's position 80ms ago, based on its speed and heading, to estimate its "current" predicted position. Then, based on the relative position and heading difference between the two vehicles, and combined with current lane information obtained from a high-precision map service, it determines whether the predicted position is on a passable path in front of the vehicle, in the same lane, or in an adjacent lane. Only data that passes this determination is sent to the subsequent fusion queue.

[0174] Step S34, the fusion of prediction and active control execution includes: First, the vehicle control system maintains a list of valid preceding vehicle data, where each entry contains the preceding vehicle ID, its spatiotemporally aligned relative position, and its... Historical sequence (last 10 seconds), its data confidence level, etc.

[0175] Then, every 200 milliseconds, the vehicle control system triggers the prediction model. The model input is a fixed-dimensional vector, composed of feature data from the 3-5 most representative vehicles in front in the list. After calculation by a 3-layer fully connected neural network, the output is a risk coefficient. .

[0176] Finally, according to The execution strategy module sends control commands to the body control module (BCM) via the CAN FD bus. For example, when At 5:00, the instruction is: "Switch the automatic wiper mode to 'Predictive Coordination Mode,' set the target frequency to 28 wipers / minute, and pre-raise the wiper arms to the standby position." After parsing this instruction, the BCM directly drives the wiper motor via a hardwired or low-voltage pulse width modulation (PWM) signal. Simultaneously, the cockpit domain controller sends graphics rendering instructions to the instrument panel and HUD via the vehicle's Ethernet network, displaying specific warning interfaces.

[0177] Step S35, during vehicle control system testing and verification: First, in the real-vehicle testing phase, organize a convoy of 5 test vehicles equipped with this vehicle control system to conduct follow-up tests in a simulated heavy rain transition zone at a closed test track. Use high-precision weather stations and vehicle-mounted cameras to record data as a real benchmark to verify the effectiveness of the early warning time.

[0178] Secondly, the vehicle control system uses bus tools to monitor and record the total end-to-end delay from receiving an emergency event broadcast from the vehicle ahead to the start of the vehicle's windshield wipers operating at a predicted frequency, ensuring that it remains consistently below the design target of 100 milliseconds.

[0179] Finally, during long-term road testing, the vehicle control system collects driver intervention data for iterative optimization of the offline training model, and periodically sends the updated model parameters to the vehicle via Over-The-Air (OTA) technology, thus completing the self-evolution of the vehicle control system.

[0180] In summary, this application provides a fully integrated, mass-producible technical solution that connects hardware infrastructure, communication protocols, core algorithms, and control execution, enabling intelligent sharing and collaborative control of vehicle-to-vehicle windshield wiper data.

[0181] Figure 4 This is a flowchart of another vehicle control method according to an embodiment of this application. Figure 4 As shown, firstly, in the data acquisition and local generation section, the vehicle sensors and controller acquire data in real time and output multiple parallel data streams. Parameter annotation boxes are then created, and data preprocessing and fusion are performed. Internal processing descriptions are then conducted to generate standardized "wiper status messages." These messages are subsequently output to V2V communication and network interconnection.

[0182] Secondly, in the V2V communication and network interconnection section, the vehicle control system makes intelligent broadcast decisions and can set multiple custom conditions for path determination. This is followed by broadcasting via the air interface.

[0183] Then, in the data reception and fusion decision-making section, the vehicle control system receives and verifies V2X messages and performs internal processing; subsequently, it aligns the spatiotemporal data and performs correlation filtering; finally, it performs multi-source data fusion and risk prediction, outputting prediction coefficients. To the section on tiered early warning and coordinated control.

[0184] Next, in the graded early warning and collaborative control section, the vehicle control system can make judgments based on preset conditions and obtain multiple decision results.

[0185] Finally, in the feedback and optimization closed-loop section, the vehicle control system will also perform vehicle control system updates. This is how the vehicle control system is updated and optimized.

[0186] According to an embodiment of this application, a vehicle control device 500 is provided. Figure 5 This is a structural block diagram of a vehicle control device according to an embodiment of this application. It should be noted that this device is applied to a target vehicle and can be used to execute the aforementioned vehicle control method. Figure 5As shown, the device includes: a receiving module 501, used to receive the preceding vehicle's operating data and the preceding vehicle's environmental perception data, wherein the preceding vehicle includes at least one connected communication vehicle located ahead of the target vehicle's driving path; a filtering module 502, used to filter the preceding vehicle's operating data and the preceding vehicle's environmental perception data according to the target vehicle's own operating data and the road topology information of the driving path, to obtain a set of valid preceding vehicle data; a prediction module 503, used to predict environmental risk information of the road segment ahead based on the preceding vehicle's operating data, the target vehicle's own operating data, and the set of valid preceding vehicle data; and a generation module 504, used to generate joint control commands for the target vehicle according to the environmental risk information, wherein the joint control commands are used to drive and control multiple actuators of the target vehicle.

[0187] Embodiments of this application also provide a vehicle, including: a memory storing an executable program; and a processor for running the program, wherein the program executes the methods described in various embodiments of this application when it runs.

[0188] Embodiments of this application also provide a computer-readable storage medium including a stored executable program, wherein, when the executable program is running, it controls the device where the computer-readable storage medium is located to perform the methods of various embodiments of this application.

[0189] Embodiments of this application also provide a computer program product, including a computer program that, when executed by a processor, implements the methods of various embodiments of this application.

[0190] Embodiments of this application also provide a computer program product, including a non-volatile computer-readable storage medium for storing a computer program that, when executed by a processor, implements the methods in various embodiments of this application.

[0191] Embodiments of this application also provide a computer program that, when executed by a processor, implements the methods described in the various embodiments of this application.

[0192] In the above embodiments of this application, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0193] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units can be a logical functional division, and in actual implementation, there may be other division methods. For instance, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling, direct coupling, or communication connection may be through some interfaces; the indirect coupling or communication connection between units or modules may be electrical or other forms.

[0194] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0195] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0196] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, read-only memory (ROM), random access memory (RAM), portable hard drive, magnetic disk, or optical disk.

[0197] The above description is only a preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A vehicle control method, characterized in that, Applied to a target vehicle, the method includes: Receives the preceding vehicle's operating data and the preceding vehicle's environmental perception data, wherein the preceding vehicle includes at least one connected communication vehicle located ahead of the target vehicle's driving path; Based on the target vehicle's own vehicle operation data and the road topology information of the driving path, the preceding vehicle's operation data and the preceding vehicle's environmental perception data are filtered and processed to obtain a set of effective preceding vehicle data. Based on the preceding vehicle's operating data, the vehicular operating data, and the set of valid preceding vehicle data, predict environmental risk information for the road section ahead. Based on the environmental risk information, a joint control command for the target vehicle is generated, wherein the joint control command is used to drive and control multiple actuators of the target vehicle.

2. The method according to claim 1, characterized in that, The preceding vehicle's operating data and the preceding vehicle's environmental perception data are obtained by parsing the wiper status messages broadcast by the preceding vehicle at a target frequency, wherein the target frequency is dynamically adjusted by the preceding vehicle according to the wiper's working mode. The wiper status message also includes a message identifier field and a data confidence field.

3. The method according to claim 2, characterized in that, The method further includes: In response to receiving an emergency event message broadcast by the vehicle ahead, the emergency event message is parsed to obtain updated vehicle ahead operation data and updated vehicle ahead environmental perception data. The emergency event message is generated by the vehicle ahead when it detects that the rate of change of the windshield wiper operating frequency within a preset time period is greater than a preset threshold. The priority of the emergency event message is higher than that of the windshield wiper status message.

4. The method according to claim 1, characterized in that, The vehicles ahead include multiple target vehicles; based on the vehicle's operating data and the road topology information, the operating data of the vehicles ahead and the environmental perception data of the vehicles ahead are filtered and processed to obtain the effective data set of the vehicles ahead, which includes: Extract the target vehicle's current position, speed, and heading angle from the vehicle's operating data; The multiple object vehicles are traversed. For the current object vehicle in the traversal, the vehicle position, vehicle speed, vehicle heading angle and message timestamp of the current object vehicle at the time of message generation are extracted from the running data of the preceding vehicle corresponding to the current object vehicle. Based on the vehicle position, vehicle speed, vehicle heading angle, and message timestamp, calculate the expected position of the current target vehicle at the current moment; Based on the road topology information, the vehicle's position, the vehicle's speed, and the vehicle's heading angle, determine whether the expected position is within the effective area corresponding to the driving path; If the expected location is within the effective area, the preceding vehicle's operating data and the preceding vehicle's environmental perception data corresponding to the current target vehicle are added to the preceding vehicle's effective data set.

5. The method according to claim 1, characterized in that, The environmental risk information includes a precipitation risk coefficient; based on the preceding vehicle's operating data, the voluntary vehicle's operating data, and the preceding vehicle's valid data set, the predicted environmental risk information includes: Based on the aforementioned set of valid data from preceding vehicles, determine the wiper operating frequency sequence and precipitation intensity estimation sequence corresponding to the valid target vehicles. Based on the preceding vehicle's operating data and the voluntary vehicle's operating data, the relative distance sequence corresponding to the valid target vehicles is determined; The wiper operating frequency sequence, the precipitation intensity estimation sequence, the relative distance sequence, and the vehicle environment parameters of the target vehicle are normalized to obtain a multi-dimensional feature vector. The multidimensional feature vector is input into the target prediction model for inference to obtain the precipitation risk coefficient. The target prediction model is a deep neural network model trained using multiple sets of sample data.

6. The method according to claim 1, characterized in that, Based on the environmental risk information, generating the joint control instructions includes: Based on the aforementioned environmental risk information, the warning level is determined; Based on the warning control strategy corresponding to the warning level, multiple sub-control commands are generated for each of the multiple actuators, wherein the multiple actuators include: wiper system, lighting system and air conditioning system.

7. The method according to claim 6, characterized in that, The early warning and control strategy includes: When the warning level is a low-risk level, a first sub-control instruction is generated, wherein the first sub-control instruction is used to control the wiper system to perform a preparation mode switching action and a wiper blade reset action. When the warning level is medium risk level, a second sub-control instruction is generated, wherein the second sub-control instruction is used to control the wiper system to perform wiping action according to the first wiping frequency determined by the environmental risk information; When the warning level is high-risk, a third sub-control instruction, a fourth sub-control instruction, and a fifth sub-control instruction are generated. The third sub-control instruction is used to control the wiper system to switch to a target continuous wiping mode, where the second wiping frequency corresponding to the target continuous wiping mode is higher than the first wiping frequency. The fourth sub-control instruction is used to control the lighting system to turn on the front and rear fog lights. The fifth sub-control instruction is used to control the air conditioning system to switch to the windshield defroster mode.

8. The method according to claim 5, characterized in that, The method further includes: After the target vehicle enters the road section ahead, the actual rainfall intensity is determined, and the intervention and control information corresponding to the target vehicle is recorded. Based on the actual precipitation intensity and the precipitation risk coefficient, the model prediction bias is determined; Based on the intervention control information and the model prediction bias, feedback sample data is constructed; Using the feedback sample data, the parameters of the target prediction model are fine-tuned to update the target prediction model.

9. A vehicle, characterized in that, include: Memory, which stores executable programs; A processor for running the program, wherein the program, when running, performs the method according to any one of claims 1 to 8.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored executable program, wherein, when the executable program is executed, it controls the device on which the storage medium is located to perform the method according to any one of claims 1 to 8.

Images