Vehicle risk early warning method and device and electronic equipment

Abstract

The invention provides a vehicle risk early warning method and device and electronic equipment, and relates to the technical field of vehicles. The method comprises the following steps: acquiring first sensing data and second sensing data of a vehicle in a driving state; the first sensing data is used for representing the motion state of the front vehicle; the second sensing data is used for representing a spatial relationship between the vehicle and surrounding vehicles; determining first risk information based on the first perception data; the first risk information is used for indicating whether the preceding vehicle is in an emergency braking risk state; determining second risk information based on the second perception data; the second risk information is used for indicating the instant relative distance between the surrounding vehicles and the vehicle and the associated collision risk level; and based on the first risk information and the second risk information, controlling the vehicle head-up display to display early warning content. The method can be used for a vehicle safety system, and can solve the problem that a vehicle driver is difficult to perceive driving risks in time.

Description

Technical Field

[0001] This application relates to the field of vehicle technology, and in particular to a vehicle risk warning method, device and electronic equipment. Background Technology

[0002] Vehicle safety warning technology is an assistance system that uses sensors such as millimeter-wave radar and cameras to detect the vehicle's surrounding environment and provide drivers with collision or lane departure risk warnings. Typical applications include forward collision warning and lane departure warning, which can effectively improve safety and reduce accident rates in scenarios such as following other vehicles.

[0003] However, existing risk warning systems are still inadequate and pose certain safety risks when faced with emergency situations such as sudden braking by the vehicle in front or sudden close proximity of a neighboring vehicle. Summary of the Invention

[0004] The purpose of this application is to provide a vehicle risk warning method, device, and electronic device, which aims to improve the real-time perception of driving risks by vehicle drivers.

[0005] Firstly, this application provides a vehicle risk warning method, the method comprising: acquiring first perception data and second perception data of a vehicle in a driving state. The first perception data characterizes the motion state of a vehicle ahead, and the second perception data characterizes the spatial relationship between the vehicle and surrounding vehicles. Based on the first perception data, first risk information is determined. The first risk information indicates whether the vehicle ahead is in an emergency braking risk state. Based on the second perception data, second risk information is determined. The second risk information indicates the instantaneous relative distance between surrounding vehicles and the associated collision risk level. Based on the first and second risk information, the vehicle's head-up display is controlled to show warning content.

[0006] The vehicle risk warning method provided in this application achieves comprehensive collection of risk elements in the traffic environment by acquiring first perception data representing the motion state of the vehicle ahead and second perception data representing the spatial relationship with surrounding vehicles, providing a complete data foundation for multi-level risk assessment. Furthermore, based on independent analysis of the first perception data, first risk information indicating the emergency braking risk state of the vehicle ahead can be obtained, accurately identifying the most common and direct rear-end collision threat. Meanwhile, based on independent analysis of the second perception data, second risk information indicating the real-time relative distance and collision risk level of surrounding vehicles can be obtained, effectively perceiving potential collaborative driving risks such as cutting in and approaching from the sides and rear. Through these two types of information, each focusing on different risk dimensions, a three-dimensional and accurate real-time risk situation map can be constructed, overcoming the blind spots or misjudgments that may exist in single-dimensional risk assessment. Finally, by using a vehicle head-up display to directly map the two types of risk information into timely visual warnings, the driver can quickly perceive key threats from different directions, effectively expanding the dimensions and timeliness of risk perception, providing the driver with more comprehensive situational awareness support, thereby enhancing the vehicle's warning capabilities and driving safety in complex traffic environments.

[0007] Optionally, the first perception data includes: first data representing the motion state of the vehicle in front, and image data of the vehicle in front; the second perception data includes: second data representing the distance between the vehicle and nearby obstacles, and point cloud data representing the three-dimensional structure of the vehicle's surrounding environment.

[0008] Optionally, acquiring the vehicle's first and second perception data while in motion includes: acquiring the relative distance and relative speed of the vehicle in front using a millimeter-wave radar deployed at the front of the vehicle, as the first data; acquiring image data of the vehicle in front using an image acquisition device deployed at the front of the vehicle; acquiring the relative distance between the positions of nearby obstacles around the vehicle and the vehicle using multiple ultrasonic sensors deployed around the vehicle, as the second data; and acquiring point cloud data of the three-dimensional structure of the vehicle's surrounding environment using multiple lidar sensors deployed around the vehicle.

[0009] Optionally, based on the first perception data, the first risk information is determined, including: based on the first data, by analyzing the relative speed change trend of the vehicle in front, the speed change characteristics of the vehicle in front are determined; image recognition is performed on the image data to determine the brake light status of the vehicle in front; based on the speed change characteristics and the brake light status, the emergency braking risk status of the vehicle in front is determined, and the first risk information is generated.

[0010] Optionally, based on speed change characteristics and brake light status, the emergency braking risk status of the vehicle ahead is determined, including: in response to the brake light status being illuminated, determining the deceleration of the vehicle ahead based on speed change characteristics; and determining that the vehicle ahead is in an emergency braking risk state when the deceleration is greater than or equal to a deceleration threshold.

[0011] Optionally, based on the second perception data, the second risk information is determined, including: identifying candidate targets around the vehicle based on point cloud data and obtaining initial information about the candidate targets; the initial information includes contour information representing the shape of the candidate targets; based on the contour information, identifying surrounding vehicles from the candidate targets; for the identified surrounding vehicles, associating the distances corresponding to the positions of the surrounding vehicles in the second data with the surrounding vehicles to determine the precise distances between the surrounding vehicles; based on the precise distances and the situation of the movement trajectories of the surrounding vehicles relative to the vehicle movement trajectories, determining the collision risk level corresponding to the surrounding vehicles as the second risk information.

[0012] Optionally, based on the precise distance of the candidate target and the state of the candidate target's trajectory relative to the vehicle's trajectory, the collision risk level corresponding to the candidate target is determined, including: predicting the relative motion trend between the candidate target and the vehicle within a preset future time period; when the relative motion trend indicates that the precise distance between the candidate target and the vehicle will be less than a safe distance threshold, the collision risk level corresponding to the candidate target is determined to be high-risk; when the relative motion trend indicates that the precise distance between the candidate target and the vehicle will be between the safe distance threshold and the warning distance threshold, the collision risk level corresponding to the candidate target is determined to be medium-risk; when the relative motion trend indicates that the precise distance between the candidate target and the vehicle will be greater than the warning distance threshold, the collision risk level corresponding to the candidate target is determined to be low-risk.

[0013] Optionally, based on the first risk information and the second risk information, the vehicle head-up display is controlled to display warning content, including: based on the first risk information indicating that the vehicle ahead is in an emergency braking risk state, the vehicle head-up display is controlled to display the first warning content corresponding to the vehicle ahead; the first warning content includes displaying a first warning icon and the speed change value of the vehicle ahead in the center area in front of the driver's field of vision; based on the second risk information, the vehicle head-up display is controlled to display the second warning content corresponding to each surrounding vehicle; the second warning content includes displaying a dynamic circular progress bar and the real-time relative distance between each surrounding vehicle in the display area corresponding to the actual position of each surrounding vehicle; the fill level of the dynamic circular progress bar is negatively correlated with the real-time relative distance; the display color of the dynamic circular progress bar is associated with the collision risk level.

[0014] Secondly, this application provides a vehicle risk warning device, comprising: a data acquisition module, a risk determination module, and a risk display module. The data acquisition module is used to acquire first perception data and second perception data of the vehicle while it is in motion. The first perception data characterizes the motion state of the vehicle ahead, and the second perception data characterizes the spatial relationship between the vehicle and surrounding vehicles. The risk determination module is used to determine first risk information based on the first perception data. The first risk information indicates whether the vehicle ahead is in an emergency braking risk state. The risk determination module is used to determine second risk information based on the second perception data. The second risk information indicates the instantaneous relative distance between surrounding vehicles and the associated collision risk level. The risk display module is used to control the vehicle's head-up display to display warning content based on the first and second risk information.

[0015] Thirdly, this application provides an electronic device comprising: a processor and a memory; the memory storing processor-executable instructions; when the processor is configured to execute the instructions, the electronic device implements the method of the first aspect described above.

[0016] Fourthly, this application provides a computer-readable storage medium comprising: computer software instructions; which, when executed in an electronic device, cause the electronic device to implement the method described in the first aspect.

[0017] Fifthly, this application provides a computer program product that, when run on a computer, causes the computer to perform the steps of the relevant method described in the first aspect above, so as to implement the method of the first aspect above.

[0018] The beneficial effects of the second to fifth aspects mentioned above can be referred to the corresponding description of the first aspect, and will not be repeated here. Attached Figure Description

[0019] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 A flowchart illustrating a vehicle risk warning method provided in an embodiment of this application; Figure 2 A flowchart illustrating a method for determining first risk information provided in an embodiment of this application; Figure 3 A flowchart illustrating a method for determining second risk information provided in an embodiment of this application; Figure 4 This is a schematic diagram illustrating the composition of a vehicle risk warning device provided in an embodiment of this application; Figure 5 This is a schematic diagram of the structure of a vehicle risk warning device provided in an embodiment of this application. Detailed Implementation

[0021] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0022] It should be noted that in the embodiments of this application, the words "exemplarily" or "for example" are used to indicate examples, illustrations, or explanations. Any embodiment or design scheme described as "exemplarily" or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design schemes. Specifically, the use of the words "exemplarily" or "for example" is intended to present the relevant concepts in a specific manner.

[0023] In the embodiments of this application, the terms "first," "second," "third," "fourth," "fifth," and "sixth" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined with "first," "second," "third," "fourth," "fifth," and "sixth" may explicitly or implicitly include one or more of that feature.

[0024] In embodiments of this application, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0025] "A and / or B" includes the following three combinations: A only, B only, and a combination of A and B.

[0026] As described in the background section, existing risk warning systems are still inadequate in emergency situations such as sudden braking by the vehicle in front or sudden close proximity of a neighboring vehicle, posing certain safety risks. Specifically: On the one hand, existing risk warning schemes have a delay from risk detection to the issuance of warnings, which directly erodes the driver's reaction time at high speeds.

[0027] On the other hand, existing risk warning schemes are limited to rough parameters such as relative distance and speed, and cannot accurately help drivers identify driving risks.

[0028] Thirdly, in existing risk warning schemes, warning information is usually displayed on the dashboard, requiring drivers to shift their attention to the road ahead to view it. This interactive design can easily lead to missed information or slow reaction in critical moments, potentially causing safety hazards.

[0029] Based on this, this application provides a vehicle risk warning method. By acquiring first perception data representing the motion state of the vehicle ahead and second perception data representing the spatial relationship with surrounding vehicles, a comprehensive collection of risk elements in the traffic environment in which the vehicle is located is achieved, providing a complete data foundation for multi-level risk assessment. Furthermore, based on the first perception data, independent analysis yields first risk information indicating the emergency braking risk state of the vehicle ahead, accurately identifying the most common and direct rear-end collision threat. Meanwhile, based on the second perception data, independent analysis yields second risk information indicating the real-time relative distance and collision risk level of surrounding vehicles, effectively perceiving potential collaborative driving risks such as cutting in and approaching from the side and rear. Through these two types of information, each focusing on different risk dimensions, a three-dimensional and accurate real-time risk situation map can be constructed, overcoming the blind spots or misjudgments that may exist in single-dimensional risk assessment. Finally, by using a vehicle head-up display to directly map the two types of risk information into timely visual warnings, the driver can quickly perceive key threats from different directions, effectively expanding the dimensions and timeliness of risk perception, providing the driver with more comprehensive situational awareness support, thereby enhancing the vehicle's warning capabilities and driving safety in complex traffic environments.

[0030] The vehicle risk warning method provided in this application is applicable to intelligent connected vehicles or vehicles with driver assistance functions that are equipped with sensor arrays, have real-time data processing capabilities, and integrate a head-up display (HUD). It can be integrated into vehicles as a vehicle risk warning system (hereinafter referred to as the system) and widely applied in various real-world driving scenarios such as highways, urban roads, and congested areas. This method is particularly suitable for situations requiring rapid judgment of emergency braking behavior of vehicles ahead and real-time collision risk assessment of surrounding vehicles, such as when following another vehicle on a highway, merging at ramps, changing lanes to overtake, or driving in adverse weather or at night where visibility or attention is limited. By fusing and analyzing the motion state of the vehicle ahead and the spatial relationship data between the vehicle and surrounding vehicles, the system can promptly and accurately generate and present tiered warning information on the head-up display, thereby effectively assisting the driver or autonomous driving system in perceiving risks in advance and making decisions, improving driving safety and experience.

[0031] The following detailed description of a vehicle risk warning method provided by this application, in conjunction with specific embodiments and accompanying drawings, provides an example of such a method.

[0032] Figure 1 This is a flowchart illustrating a vehicle risk warning method provided in an embodiment of this application. Specifically, as shown... Figure 1 As shown, it includes the following: S101. Obtain the first and second perception data of the vehicle in the driving state.

[0033] The first perception data is used to characterize the motion state of the vehicle in front, and the second perception data is used to characterize the spatial relationship between the vehicle and surrounding vehicles.

[0034] In this embodiment, the first perception data is a comprehensive dataset used to accurately characterize the motion state of the vehicle ahead. It primarily includes two types of key information: one type is first data that quantifies the motion characteristics of the vehicle ahead (such as relative distance and relative speed), and the other type is image data that captures the visual features of the vehicle ahead (such as vehicle body outline and brake light status). The second perception data is a dataset used to clearly characterize the spatial relationship between the vehicle and surrounding vehicles. It mainly includes second data characterizing the relative distance between the vehicle and nearby obstacles, as well as point cloud data that presents the three-dimensional structure of the vehicle's surrounding environment. Through the complementarity of these two types of data, the spatial information such as the position and distance of surrounding vehicles can be comprehensively reflected.

[0035] In some embodiments, the system can acquire first and second perception data under various driving conditions, such as high-precision millimeter-wave radar deployed at the front of the vehicle, a front-end image acquisition device, and multiple ultrasonic sensors and multiple lidars deployed around the vehicle in the four directions (front, rear, left, and right). Specifically, the system can: acquire the relative distance and relative speed of vehicles ahead using the front-end millimeter-wave radar, as the first data in the first perception data; capture real-time images of the road scene ahead using the front-end image acquisition device, acquiring image data including vehicles ahead, as another part of the first perception data; emit ultrasonic pulses and receive echoes using the multiple surrounding ultrasonic sensors, acquiring the relative distances between the vehicle and nearby obstacles around the vehicle, as the second data in the second perception data; and emit laser beams to scan the surrounding environment using the multiple surrounding lidars, receiving reflected light signals and generating three-dimensional point cloud data, as another part of the second perception data.

[0036] In some embodiments, during the acquisition of first and second sensing data through various sensors, the system may also: perform real-time preliminary filtering on the relative distance and relative speed data acquired by the millimeter-wave radar to remove abnormal fluctuations caused by environmental interference, ensuring the stability of the first data; perform real-time grayscale preprocessing on the image data acquired by the image acquisition device to quickly highlight the outline of the vehicle ahead and key areas such as brake lights in the image, reducing the amount of data processing for subsequent image recognition; perform validity verification on the distance data acquired by the ultrasonic sensor, removing invalid data that exceeds its effective detection range; and perform preliminary noise reduction processing on the three-dimensional point cloud data generated by the lidar to remove interference from environmental noise points on the identification of effective targets, while recording the acquisition timestamps of each sensor to ensure the time synchronization of different types of sensing data.

[0037] In one possible implementation, after the vehicle starts and enters driving mode, the system activates the deployed sensors and puts them into real-time operation. The front-end millimeter-wave radar operates on a preset millimeter-wave frequency band, continuously emitting high-frequency electromagnetic waves at a fixed frequency. When the electromagnetic waves encounter a vehicle ahead, they are reflected. The radar receives the reflected signal and, based on the Doppler effect, calculates the signal frequency change to output the relative distance and relative speed data between the vehicle ahead and the vehicle in real time, storing this data in a designated data area as the first data. A front-end high-resolution camera, working in conjunction with the millimeter-wave radar, continuously captures images of the road ahead at a preset high frame rate (e.g., 30 frames per second). The image data is uploaded to the system in real-time via a transmission link, serving as image data in the first perception data. Ultrasonic sensors deployed around the front, rear, left, and right of the vehicle sequentially emit ultrasonic pulses at preset time intervals. When the ultrasonic waves propagate through the air and encounter obstacles such as surrounding vehicles, they are reflected to form echoes. After receiving the echoes, the sensors calculate the relative distance between the vehicle and the corresponding obstacle based on the fixed propagation speed and round-trip time of the ultrasonic waves. The measurement data from all ultrasonic sensors are summarized as the second data. The surrounding LiDAR continuously emits laser beams at a set scanning frequency, covering a 360-degree area around the vehicle. When the laser beams encounter surrounding vehicles, road facilities, or other targets, they are reflected back. The LiDAR receives the reflected light signals and records information such as propagation time and angle. It then generates 3D point cloud data of the vehicle's surrounding environment through a 3D modeling algorithm. This data is transmitted to the system in real time as part of the second perception data. Finally, the system integrates the above data to form complete first and second perception data.

[0038] As can be seen from step S101, the system achieves comprehensive acquisition of first and second perception data through multi-sensor collaborative acquisition. The first perception data simultaneously covers the quantitative motion data (relative distance and speed) and visual state data (image) of the vehicle in front, while the second perception data takes into account the distance measurement data of the surrounding near distance and the three-dimensional structural data of the far distance. This effectively makes up for the limitations of a single sensor in terms of data acquisition range, accuracy and dimensionality, and ensures the comprehensiveness and accuracy of the perception data.

[0039] S102. Based on the first perception data, determine the first risk information.

[0040] The first risk information is used to indicate whether the vehicle in front is in a state of emergency braking risk.

[0041] In this embodiment, the first risk information is key information derived from the analysis of the first perception data, used to clearly indicate whether the vehicle ahead is in an emergency braking risk state. Its core function is to provide the driver with an accurate basis for judging the braking risk of the vehicle ahead, avoiding untimely response due to information lag or misjudgment. An emergency braking risk state refers to a driving state in which the speed of the vehicle ahead drops sharply in a short period of time, accompanied by characteristics such as the brake lights illuminating, which may pose a direct threat to the driving safety of the following vehicle. The determination of this state requires a comprehensive judgment based on vehicle motion data and visual feature data.

[0042] In some embodiments, the system can extract first data (such as relative distance and relative speed) and image data of the vehicle in front by calling the first perception data collected and stored in step S101. In driving scenarios where the vehicle is traveling at high speed or following another vehicle on urban roads, which require key monitoring of the vehicle in front, the system can obtain basic data for determining the first risk information.

[0043] In one possible implementation, the specific implementation of step S102 is described below. Figure 2 The description of the embodiments shown is not detailed here.

[0044] As shown in step S102, the system, based on the comprehensive and accurate first perception data obtained in step S101, focuses on the core requirement of whether the vehicle in front is in an emergency braking risk state. Through targeted data analysis and processing, it generates first risk information, effectively avoiding the risk misjudgment problem caused by relying on only a single data dimension. The first risk information directly clarifies the key dangerous state of the vehicle in front, laying the core decision-making foundation for providing timely and effective warnings to the driver via the vehicle's HUD. It also precisely meets the actual need for rapid identification of the emergency braking risk of the vehicle in front in driving scenarios, making up for the shortcomings of existing technologies in comprehensively judging the speed change trend and braking status of the vehicle in front.

[0045] S103. Based on the second perception data, determine the second risk information.

[0046] The second risk information is used to indicate the real-time relative distance between surrounding vehicles and the associated collision risk level.

[0047] In this embodiment, the second risk information refers to comprehensive information formed after analysis and processing of data related to the spatial relationship between the vehicle and surrounding vehicles. This information clearly indicates the instantaneous relative distance between the vehicle and surrounding vehicles, as well as the collision risk level associated with that distance. Its core value lies in providing the driver with a quantitative reference for the degree of danger posed by surrounding vehicles. The instantaneous relative distance refers to the actual straight-line distance between the vehicle and surrounding vehicles at a certain moment, reflecting the degree of spatial proximity between them in real time. The collision risk level is a risk hierarchy divided according to factors such as the instantaneous relative distance and trajectory of surrounding vehicles and the vehicle. It is used to intuitively reflect the likelihood of a collision and is an important basis for setting the subsequent warning intensity.

[0048] In some embodiments, the system can extract second data representing the distance between the vehicle and nearby obstacles and point cloud data representing the three-dimensional structure of the vehicle's surrounding environment from the second perception data collected and stored in step S101. In driving scenarios where the vehicle is in urban traffic congestion, highway lane changes, roundabouts, or other situations where there are many surrounding vehicles or where the surrounding environment needs to be frequently monitored, the system can obtain basic data for determining the second risk information.

[0049] In one possible implementation, the specific implementation of step S103 is described below. Figure 3 The description of the embodiments shown is not detailed here.

[0050] As shown in step S103, the system, supported by the multi-dimensional, high-precision second perception data obtained in step S101, focuses on two core pieces of information: the real-time relative distance between the vehicle and surrounding vehicles and the collision risk level. Through targeted data analysis and processing, it generates second risk information, effectively addressing the shortcomings of existing technologies that struggle to accurately obtain close-up details of surrounding vehicles and quantify collision risks. The second risk information not only clarifies specific distance data but also intuitively presents the degree of danger through risk levels, providing a precise decision-making basis for the subsequent targeted display of warning content on the vehicle's HUD. Simultaneously, it accurately matches the actual need for rapid perception of dynamic risks from surrounding vehicles during driving, helping drivers to anticipate potential threats from surrounding vehicles and allowing sufficient reaction time to avoid scratches and collisions.

[0051] S104. Based on the first risk information and the second risk information, control the vehicle's head-up display to show the warning content.

[0052] In this embodiment, the warning content refers to comprehensive alert information generated based on first and second risk information, used to intuitively convey the dangerous state of the vehicle's surroundings to the driver. It encompasses visual alert elements and associated auxiliary alerts, and its core function is to allow the driver to quickly understand the type and severity of the risk, as well as relevant key data. A vehicle head-up display (HUD) is a display device that projects the warning content onto the area directly in front of the driver's field of vision, allowing the driver to obtain information without looking down at other displays, thus minimizing distraction.

[0053] In some embodiments, the system can obtain the core risk data required to generate warning content when the vehicle detects an emergency braking risk of the vehicle in front, a collision risk of surrounding vehicles (or a high / medium risk level), or both types of risks during vehicle operation by calling the first risk information determined in step S102 and the second risk information determined in step S103, thus providing a basis for subsequent HUD display control.

[0054] In some embodiments, during the process of controlling the HUD display of warning content based on the first risk information and the second risk information, the system can also: adjust the display intensity of the first warning content according to the urgency of the first risk information, such as increasing the flashing frequency of the red flashing icon when the vehicle in front brakes suddenly as the deceleration increases, and enlarging the font of the speed change value as the risk level increases. For the second risk information, when the collision risk level of surrounding vehicles upgrades to high risk, a red border is added to the flashing dynamic circular progress bar, and the vehicle's auditory warning device is simultaneously triggered, emitting a rapid beeping sound matching the risk level (the higher the risk level, the faster the beeping frequency). If both the first and second risk information are detected simultaneously, the system prioritizes placing the first warning content in the most prominent position on the HUD display interface, while the second warning content is distributed in the corresponding positions without obstructing the core field of vision, avoiding visual interference caused by multiple types of warning information.

[0055] In one possible implementation, the system can: First, prioritize the acquired first and second risk information, with the first risk information (risk of emergency braking by the vehicle ahead) having higher priority than the second risk information (risk of collision with surrounding vehicles). When the first risk information indicates that the vehicle ahead is in a state of emergency braking risk, the system controls the HUD to display the first warning content in the center area in front of the driver's field of vision. Specifically, this is a red flashing icon that flashes continuously at a preset frequency, with a white value showing the speed change of the vehicle ahead displayed to the right of the icon. The value is set to a size that is easy to read quickly, ensuring that the driver can immediately capture the core speed change data. When the second risk information indicates that there is a risk of collision with surrounding vehicles, the system displays the second warning content in the corresponding display area of ​​the HUD according to the actual position of the surrounding vehicles recorded in the second risk information. Specifically, this is a dynamic circular progress bar. The degree of filling of the progress bar is negatively correlated with the real-time relative distance between the surrounding vehicles and the vehicle (the closer the distance, the higher the filling degree). The display color of the progress bar is associated with the collision risk level (green corresponds to low risk level, yellow corresponds to medium risk level, and red corresponds to high risk level). At the same time, a conspicuous real-time relative distance number is marked next to the circular progress bar. When two types of risk information exist simultaneously, the first warning content is displayed in the center of the field of vision, while the second warning content is displayed in a semi-transparent overlay in the corresponding surrounding area to avoid obscuring the first warning content and ensure that both types of risk information are clearly presented without interfering with the driver's observation of the road ahead.

[0056] As shown in step S104, the system presents warning content through a HUD (Head-Up Display), a device positioned within the driver's field of vision. This solves the problem of traditional warning information on the dashboard requiring the driver to look down, minimizing driver distraction and ensuring that risk warnings are detected immediately. Different warning content formats are designed for different risk types. The first warning content features a flashing red icon and speed value, while the second warning content uses a dynamic circular progress bar and distance figures. This provides a visually intuitive and quantifiable presentation of risk information, allowing drivers to quickly understand the core risk data and make decisions without additional analysis. Through priority determination, display location planning, and additional intensity adjustment and auditory assistance functions, the system ensures comprehensive display of various risk information while avoiding visual interference, further enhancing the effectiveness and practicality of warnings. This provides timely and accurate information support for drivers to quickly respond to emergencies such as sudden braking by the vehicle in front or approaching vehicles, effectively reducing the probability of accidents.

[0057] In this embodiment of the application, in step S102 above, the system can determine the first risk information by calling the first sensing data collected and stored in step S101, and combining the first data with image data. Specifically, as shown... Figure 2 As shown, step S102 above can be specifically implemented as S201-S203: S201. Based on the first data, determine the speed change characteristics of the vehicles ahead by analyzing the relative speed change trend of the vehicles ahead.

[0058] In this embodiment, the relative speed change trend refers to the direction of the speed change of the vehicle in front relative to the vehicle itself over a continuous time period, reflecting whether the speed is stable, decreasing slowly, or decreasing sharply. Speed ​​change characteristics are core attributes extracted after analyzing the relative speed change trend, such as key parameters like the rate, magnitude, and duration of speed decrease.

[0059] S202. Perform image recognition on the image data to determine the status of the brake lights of the vehicle ahead.

[0060] In this embodiment of the application, the brake light status refers to the working status of the brake lights of the vehicle in front, specifically divided into the illuminated state and the unilluminated state, which is a direct visual basis for judging whether the vehicle is braking.

[0061] S203. Based on speed change characteristics and brake light status, determine the emergency braking risk status of the vehicle ahead and generate first risk information.

[0062] In this embodiment of the application, the emergency braking risk state refers to a dangerous state in which the vehicle in front may cause a rear-end collision due to a sudden decrease in speed and the illumination of its brake lights.

[0063] In some embodiments, during the process of determining the emergency braking risk state of the vehicle ahead based on speed change characteristics and brake light status, the system can: dynamically adjust the deceleration threshold for different driving scenarios, such as setting a relatively low deceleration threshold in high-speed driving scenarios (because even a small sudden decrease in speed at high speeds can trigger a high risk), and appropriately increasing the deceleration threshold in low-speed urban driving scenarios, thereby improving the adaptability of the judgment in different scenarios. The system performs secondary verification on the brake light status recognition results, analyzing the consistency of brightness changes in the brake light area across multiple consecutive frames to eliminate misidentification caused by ambient light (such as strong light or streetlight reflection), ensuring the accuracy of brake light status determination. When the brake light status is detected as illuminated but the deceleration has not reached the threshold, auxiliary prompt information indicating normal braking of the vehicle ahead is generated, distinguishing it from the emergency braking risk state and providing the driver with a more detailed reference to the vehicle ahead's status. Simultaneously, the system records the speed change characteristic data and brake light status information for each determination, for subsequent algorithm optimization and iteration, improving the determination accuracy.

[0064] In one possible implementation, the system first processes the acquired initial data into a time series. A sliding window algorithm is used to select relative speed data from five consecutive acquisition cycles. By calculating the difference in relative speed between adjacent cycles, the relative speed change rate is obtained. Then, a linear fitting algorithm is used to analyze the trend of this change rate, extracting parameters such as the deceleration rate of the vehicle ahead and the magnitude of speed change per unit time, forming speed change features. Simultaneously, the system preprocesses the image data, including grayscale conversion, Gaussian noise reduction, and edge detection, highlighting the rear region of the vehicle ahead in the image. A pre-trained deep learning image recognition model (trained based on a large number of sample images with and without illuminated brake lights) is then called to perform feature matching on the preprocessed image, identifying the brightness, color, and illuminated area of ​​the brake lights of the vehicle ahead, and outputting whether the brake lights are illuminated or not. Finally, the system determines whether the brake lights are illuminated. If they are not illuminated, it directly determines that the vehicle ahead is not in an emergency braking risk state, generating a first risk information indicating no emergency braking risk. If the indicator is lit, the calculated speed change characteristics are combined with relative speed data to calculate the actual deceleration of the vehicle ahead (obtained by the ratio of speed change to time interval). The calculated deceleration is compared with a preset deceleration threshold (e.g., 8 m / s²). When the deceleration is greater than or equal to the threshold, the vehicle ahead is determined to be in an emergency braking risk state, generating the first risk information of emergency braking risk. At the same time, the deceleration value and the basis for the judgment are recorded to provide data support for the generation of subsequent warning content.

[0065] As shown in steps S201-S203, the system effectively avoids misjudgment caused by relying solely on speed or visual data by employing a dual judgment logic of quantified speed analysis and visual state recognition, thus improving the accuracy of emergency braking risk state judgment. By extracting speed change features and calculating deceleration, combined with preset thresholds for judgment, the recognition criteria for emergency braking become clearer and more quantifiable, solving the deficiency of ambiguous judgment of the preceding vehicle's emergency braking in existing technologies. Preprocessing and recognition of image data ensures the accuracy of brake light status recognition and reduces the impact of environmental interference. Ultimately, the first risk information generated by the system accurately reflects the emergency braking risk of the preceding vehicle, providing a reliable and accurate decision-making basis for subsequent HUD warnings, ensuring that the driver receives a true and effective hazard warning, and allowing sufficient reaction time to avoid rear-end collisions.

[0066] In this embodiment of the application, in step S103 above, the system can determine the second risk information by calling the second sensing data collected and stored in step S101, and combining the second data with the point cloud data of the surrounding environment. Specifically, as shown... Figure 3 As shown, step S103 above can be specifically implemented as S301-S304: S301. Based on point cloud data, identify candidate targets around the vehicle and obtain initial information about the candidate targets.

[0067] The initial information includes contour information that characterizes the shape of the candidate target.

[0068] In this embodiment, candidate targets are surrounding objects identified from point cloud data that may affect the vehicle's movement, covering various obstacles such as vehicles, road facilities, and pedestrians. Initial information is basic descriptive data of candidate targets, the core of which includes contour information, i.e., feature data that reflects the outline of the candidate target, which is the key basis for distinguishing target types.

[0069] S302. Based on contour information, identify surrounding vehicles from candidate targets.

[0070] In this embodiment of the application, the surrounding vehicles are other vehicles that may interact with the vehicle in space, selected from the candidate targets.

[0071] S303. For the identified surrounding vehicles, associate the distances corresponding to the positions of the surrounding vehicles in the second data with the surrounding vehicles to determine the precise distances between the surrounding vehicles.

[0072] In this embodiment of the application, the precise distance refers to the actual real-time distance between the vehicle and surrounding vehicles after data association and calibration, which has higher accuracy than the original collected data.

[0073] S304. Based on the precise distance and the relative status of the movement trajectories of surrounding vehicles to the vehicle's movement trajectory, determine the corresponding collision risk level of the surrounding vehicles as the second risk information.

[0074] In this embodiment, the motion trajectory situation refers to the relative relationship between the motion trajectories of surrounding vehicles and the motion trajectory of the vehicle itself, including whether there is an intersection or a tendency to approach. The collision risk level is divided based on the precise distance and motion trajectory situation, and is used to quantify the level (including high, medium, and low) of the probability of a collision between surrounding vehicles and the vehicle itself.

[0075] In some embodiments, during the process of determining the collision risk level of surrounding vehicles based on precise distance and motion trajectory, the system can: preset differentiated safe distance thresholds and warning distance thresholds according to different driving scenarios (such as highways and urban low-speed roads). The thresholds are set larger in highway scenarios to adapt to the braking requirements of high-speed driving, while the thresholds are appropriately reduced in urban low-speed scenarios to avoid over-warning. The system smooths the predicted motion trajectories of surrounding vehicles and the vehicle itself, using a Kalman filter algorithm to correct fluctuations in the trajectory data and improve the stability of motion trend prediction. Candidate targets are continuously tracked; if a candidate target is identified as a surrounding vehicle for multiple consecutive acquisition cycles and its motion trajectory is stable, its data processing priority is increased to ensure the timeliness of risk level determination. Simultaneously, when the collision risk level upgrades from low to medium or high, the system automatically records data such as precise distance changes and motion trajectory intersections during this process, providing sample support for subsequent threshold optimization algorithms.

[0076] In one possible implementation, the system first preprocesses the point cloud data, including noise removal and coordinate calibration. Then, it uses an Euclidean distance-based clustering algorithm to group the preprocessed point cloud data, clustering points with similar spatial locations into independent candidate targets. Initial information for the candidate targets is generated by extracting the boundary coordinates and shape contour features of each cluster region, with a focus on retaining contour information for subsequent filtering. Next, the system calls a pre-trained vehicle contour recognition model (trained based on contour samples from a large number of different types of vehicles) to match the contour information of each candidate target with the vehicle contour features in the model. If the matching degree exceeds a preset threshold (e.g., 85%), the candidate target is determined to be a nearby vehicle, completing the filtering from candidate targets to nearby vehicles. Subsequently, the system can use the determined azimuth information of nearby vehicles in the point cloud data, correlate it with distance data collected by ultrasonic sensors in the corresponding azimuth in the second set of data, and combine it with the three-dimensional coordinates of nearby vehicles in the lidar point cloud data. Using triangulation principles, the two sets of data are fused and calibrated to eliminate sensor errors, and the precise distance between the nearby vehicles and the current vehicle is calculated. Then, based on the historical location data and real-time speed and direction information of surrounding vehicles, the system uses a multinomial fitting algorithm to predict their trajectory within a preset time period (e.g., 3 seconds). Simultaneously, it combines this with the vehicle's own driving plan (based on current speed, turn signals, etc.) to determine its trajectory, analyzing the trajectory dynamics of both. The predicted future relative distance is compared with preset safe distance thresholds and warning distance thresholds. If the future relative distance is less than the safe distance threshold, the collision risk level is determined to be high. If the future relative distance is between the safe distance threshold and the warning distance threshold, it is determined to be medium risk. If the future relative distance is greater than the warning distance threshold, it is determined to be low risk. Finally, the system can integrate the precise distances of all surrounding vehicles with their corresponding collision risk levels to generate secondary risk information.

[0077] As shown in steps S301-S304, the system achieves accurate identification and precise distance calculation of surrounding vehicles through the collaborative use of point cloud data and secondary data, solving the problems of insufficient accuracy of single sensor data and fuzzy target recognition. By filtering surrounding vehicles using contour information, the system avoids misjudging non-vehicle targets such as road facilities and pedestrians as risk sources, improving the targeted nature of risk assessment. Predicting future relative motion trends based on motion trajectory and combining differentiated thresholds to classify collision risk levels makes risk assessment more aligned with actual driving scenarios. The quantified risk levels allow drivers to intuitively understand the degree of danger. Ultimately, the secondary risk information generated by the system is both accurate and practical, providing a reliable basis for subsequent targeted display of warning content on the HUD, helping drivers to anticipate potential collision risks from surrounding vehicles, providing sufficient reaction time to avoid scratches and collisions, and improving driving safety.

[0078] The vehicle risk warning method provided in this application achieves comprehensive collection of risk elements in the traffic environment by acquiring first perception data representing the motion state of the vehicle ahead and second perception data representing the spatial relationship with surrounding vehicles, providing a complete data foundation for multi-level risk assessment. Furthermore, based on independent analysis of the first perception data, first risk information indicating the emergency braking risk state of the vehicle ahead can be obtained, accurately identifying the most common and direct rear-end collision threat. Meanwhile, based on independent analysis of the second perception data, second risk information indicating the real-time relative distance and collision risk level of surrounding vehicles can be obtained, effectively perceiving potential collaborative driving risks such as cutting in and approaching from the sides and rear. Through these two types of information, each focusing on different risk dimensions, a three-dimensional and accurate real-time risk situation map can be constructed, overcoming the blind spots or misjudgments that may exist in single-dimensional risk assessment. Finally, by using a vehicle head-up display to directly map the two types of risk information into timely visual warnings, the driver can quickly perceive key threats from different directions, effectively expanding the dimensions and timeliness of risk perception, providing the driver with more comprehensive situational awareness support, thereby enhancing the vehicle's warning capabilities and driving safety in complex traffic environments.

[0079] In an exemplary embodiment, Figure 4 This is a schematic diagram illustrating the composition of a vehicle risk warning device provided in an embodiment of this application. Figure 4 As shown, the vehicle risk warning device includes: a data acquisition module 401, a risk determination module 402, and a risk display module 403. The data acquisition module 401 acquires first and second perception data of the vehicle while it is in motion. The first perception data characterizes the motion state of the vehicle ahead, and the second perception data characterizes the spatial relationship between the vehicle and surrounding vehicles. The risk determination module 402 determines first risk information based on the first perception data. The first risk information indicates whether the vehicle ahead is in an emergency braking risk state. The risk determination module 402 also determines second risk information based on the second perception data. The second risk information indicates the instantaneous relative distance between surrounding vehicles and the associated collision risk level. The risk display module 403 controls the vehicle's head-up display to show the warning content based on the first and second risk information.

[0080] In an exemplary embodiment, this application also provides an electronic device, which may be the vehicle risk warning device in the above method embodiments. Figure 5 This is a schematic diagram of the structure of a vehicle risk warning device provided in an embodiment of this application. Figure 5As shown, the vehicle risk warning device may include: a processor 501 and a memory 502; the memory 502 stores instructions executable by the processor 501; when the processor 501 is configured to execute instructions, it causes the electronic device or network device or manager to perform the system functions described in the foregoing method embodiments.

[0081] In exemplary embodiments, this application also provides a computer-readable storage medium storing computer program instructions thereon; when the computer program instructions are executed by a computer, the computer causes the computer to implement the methods described in the foregoing embodiments. The computer may be an electronic device, a network device, or a manager. The computer-readable storage medium may be a non-transitory computer-readable storage medium, such as a ROM, random access memory (RAM), CD-ROM, magnetic tape, floppy disk, and optical data storage device.

[0082] In an exemplary embodiment, this application also provides a computer program product that, when run on a computer, causes the computer to execute the aforementioned related method steps to implement the service recommendation method described above.

[0083] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A vehicle risk warning method, characterized in that, The method includes: Acquire first and second perception data of the vehicle while it is in motion; the first perception data is used to characterize the motion state of the vehicle ahead; the second perception data is used to characterize the spatial relationship between the vehicle and surrounding vehicles. Based on the first sensing data, a first risk information is determined; the first risk information is used to indicate whether the vehicle in front is in an emergency braking risk state. Based on the second perception data, second risk information is determined; the second risk information is used to indicate the instantaneous relative distance between the surrounding vehicles and the vehicle and the associated collision risk level; Based on the first risk information and the second risk information, the vehicle's head-up display is controlled to show the warning content.

2. The method according to claim 1, characterized in that, The first perception data includes: first data characterizing the motion state of the vehicle ahead, and image data of the vehicle ahead; The second perception data includes: second data representing the distance between the vehicle and nearby obstacles, and point cloud data representing the three-dimensional structure of the vehicle's surrounding environment.

3. The method according to claim 2, characterized in that, The acquisition of the vehicle's first and second perception data under driving conditions includes: The relative distance and relative speed of the vehicle in front are obtained by a millimeter-wave radar deployed at the front of the vehicle, and used as the first data; The image data of the vehicle in front is acquired by an image acquisition device deployed at the front of the vehicle; The relative distances between the positions of nearby obstacles around the vehicle and the vehicle are obtained by multiple ultrasonic sensors deployed around the vehicle, and this is used as the second data. The point cloud data of the three-dimensional structure of the vehicle's surrounding environment is acquired by multiple lidar sensors deployed around the vehicle.

4. The method according to claim 2, characterized in that, The step of determining the first risk information based on the first perceived data includes: Based on the first data, the speed change characteristics of the vehicles ahead are determined by analyzing the relative speed change trend of the vehicles ahead. Image recognition is performed on the image data to determine the brake light status of the vehicle ahead; Based on the speed change characteristics and the brake light status, the emergency braking risk status of the vehicle ahead is determined, and the first risk information is generated.

5. The method according to claim 4, characterized in that, Determining the emergency braking risk status of the vehicle ahead based on the speed change characteristics and the brake light status includes: In response to the brake light being illuminated, the deceleration of the vehicle ahead is determined based on the speed change characteristics; When the deceleration is greater than or equal to the deceleration threshold, the vehicle ahead is determined to be in the emergency braking risk state.

6. The method according to claim 2, characterized in that, The determination of the second risk information based on the second perceived data includes: Based on the point cloud data, candidate targets existing around the vehicle are identified, and initial information of the candidate targets is obtained; the initial information includes contour information characterizing the shape of the candidate targets. Based on the contour information, surrounding vehicles are determined from the candidate targets; For the identified surrounding vehicles, the distances in the second data corresponding to the positions of the surrounding vehicles are associated with the surrounding vehicles to determine the precise distances between the surrounding vehicles and the vehicle. Based on the precise distance and the relative position of the movement trajectories of the surrounding vehicles to the vehicle's movement trajectory, the collision risk level corresponding to the surrounding vehicles is determined as the second risk information.

7. The method according to claim 6, characterized in that, The method of determining the collision risk level corresponding to the candidate target based on the precise distance of the candidate target and the situation of the candidate target's trajectory relative to the vehicle's trajectory includes: Predict the relative motion trend between the candidate target and the vehicle over a future preset time period; When the relative motion trend indicates that the precise distance between the candidate target and the vehicle will be less than the safe distance threshold, the collision risk level corresponding to the candidate target is determined to be a high-risk level. When the relative motion trend indicates that the precise distance between the candidate target and the vehicle will be between the safe distance threshold and the warning distance threshold, the collision risk level corresponding to the candidate target is determined to be a medium risk level. When the relative motion trend indicates that the precise distance between the candidate target and the vehicle will be greater than the warning distance threshold, the collision risk level corresponding to the candidate target is determined to be a low-risk level.

8. The method according to any one of claims 1-7, characterized in that, The step of controlling the vehicle's head-up display to show warning content based on the first risk information and the second risk information includes: Based on the first risk information indicating that the vehicle ahead is in a state of emergency braking risk, the vehicle head-up display is controlled to display a first warning content corresponding to the vehicle ahead; the first warning content includes displaying a first warning icon and the speed change value of the vehicle ahead in the center area of ​​the driver's field of vision. Based on the second risk information, the vehicle head-up display is controlled to display second warning content corresponding to each surrounding vehicle; the second warning content includes a display area corresponding to the actual position of each surrounding vehicle, displaying a dynamic circular progress bar and the instantaneous relative distance between each surrounding vehicle and the vehicle; the fill level of the dynamic circular progress bar is negatively correlated with the instantaneous relative distance; the display color of the dynamic circular progress bar is associated with the collision risk level.

9. A vehicle risk warning device, characterized in that, The device includes: a data acquisition module, a risk determination module, and a risk display module; The data acquisition module is used to acquire first perception data and second perception data of the vehicle in a driving state; the first perception data is used to characterize the motion state of the vehicle in front; the second perception data is used to characterize the spatial relationship between the vehicle and surrounding vehicles. The risk determination module is used to determine first risk information based on the first sensing data; the first risk information is used to indicate whether the vehicle in front is in an emergency braking risk state. The risk determination module is used to determine second risk information based on the second perception data; the second risk information is used to indicate the instantaneous relative distance between the surrounding vehicles and the vehicle and the associated collision risk level; The risk display module is used to control the vehicle head-up display to display warning content based on the first risk information and the second risk information.

10. An electronic device, characterized in that, The electronic device includes: a processor and a memory; The memory stores instructions that the processor can execute; When the processor is configured to execute the instructions, the electronic device performs the method as described in any one of claims 1-8.

Images