Vehicle collision risk warning and accident warning method

Abstract

The present invention discloses a vehicle collision risk warning and accident warning method, which relates to the technical field of vehicle collision risk warning, and includes the following steps: installing a multimodal vehicle-mounted sensor in a vehicle; collecting driving data and environmental data of the vehicle in real time during driving through the multimodal vehicle-mounted sensor; collecting the driver's braking habits; customizing different risk limits for the vehicle according to the braking habits and environmental data; judging the collision accident risk of the vehicle in real time based on the driving data, environmental data and risk limits, and obtaining a collision accident risk level; and issuing a risk warning to the driver based on the collision accident risk level. The present invention is used to solve the problem that the existing vehicle collision risk warning technology still has the problem of issuing a warning for vehicle collision risk by means of fixed thresholds or fixed usage scenarios, resulting in high restrictions on usage scenarios and an inability to ensure that the vehicle has sufficient braking space.

Description

Technical Field

[0001] The present invention relates to the technical field of vehicle collision risk warning, and in particular to a vehicle collision risk warning and accident warning method. Background Art

[0002] Vehicle collision risk warning technology refers to the use of on-board sensors, V2X vehicle networking and environmental perception systems to collect real-time vehicle motion status, dynamic information of surrounding targets and road conditions, calculate the collision probability and urgency through algorithm models, and provide early warning to the driver.

[0003] Existing vehicle collision risk warning technologies usually use fixed threshold pairs and only issue warnings when the collision time is very short. They cannot effectively predict the driver in advance, and different drivers have different driving habits, including their habits when stepping on the brake pedal. Existing vehicle collision risk warning technologies can usually only remind drivers to slow down in emergency situations, or remind drivers to slow down in fixed scenarios, such as when entering school sections and sidewalks, regardless of whether there are pedestrians in front, they will issue a slowdown reminder. In addition, when reminding drivers to slow down in emergency situations, it is usually when the collision time is short, and the driver needs to step on the brakes or AEB intervene to avoid it. Collision avoidance cannot set different safety distances for vehicles according to different driving scenarios to ensure that the vehicle has sufficient braking space. For example, in the patent application with publication number CN119149406A, a "vehicle collision risk warning method, device and storage medium in a simulation scenario" is disclosed. This solution determines whether there is a collision risk between vehicles by presetting a collision time threshold. The usage scenario is not flexible enough and the effectiveness is low. The existing vehicle collision risk warning technology also has the problem of warning vehicle collision risks through fixed thresholds or fixed usage scenarios, resulting in high restrictions on usage scenarios and an inability to ensure that the vehicle has sufficient braking space. Summary of the Invention

[0004] The present invention aims to solve one of the technical problems in the prior art to at least a certain extent, by installing a multimodal vehicle-mounted sensor in the vehicle and collecting driving data and environmental data of the vehicle in real time during driving through the multimodal vehicle-mounted sensor, then collecting the driver's braking habits, and then analyzing the duration of the user's commonly used braking scenarios based on the braking habits to obtain the commonly used braking duration, and then customizing different risk limits for the vehicle based on the commonly used braking duration and environmental data, and then judging the vehicle's collision accident risk in real time based on the driving data, environmental data and risk limits to obtain the collision accident risk level, and finally issuing a risk warning to the driver based on the collision accident risk level, so as to solve the problem that the existing vehicle collision risk warning technology still has the problem of warning the vehicle collision risk through fixed thresholds or fixed usage scenarios, resulting in high restrictions on usage scenarios and failure to ensure that the vehicle has sufficient braking space.

[0005] To achieve the above objectives, the present application provides a vehicle collision risk warning and accident warning method, comprising the following steps:

[0006] Installing multimodal onboard sensors in vehicles;

[0007] The multimodal vehicle-mounted sensor collects driving data and environmental data of the vehicle in real time during driving;

[0008] Collect drivers' braking habits;

[0009] Customize different risk limits for vehicles based on braking habits and environmental data;

[0010] Based on driving data, environmental data and risk limits, the collision risk of the vehicle is judged in real time to obtain the collision risk level;

[0011] Provide risk warning to drivers based on the collision accident risk level.

[0012] Furthermore, a multimodal vehicle-mounted sensor is installed in the vehicle, and the multimodal vehicle-mounted sensor includes a laser radar, a high-definition camera, and a friction coefficient sensor.

[0013] Furthermore, the multimodal vehicle-mounted sensor collects driving data and environmental data in real time during the vehicle's driving process. The driving data is specifically 3DHMI data collected by a lidar and a high-definition camera, and the environmental data is specifically the road friction coefficient detected by a friction coefficient sensor. The 3DHMI data is used to perform 3D modeling and display of pedestrians, vehicles and objects around the vehicle, and at the same time mark the distance between the vehicle and each pedestrian, vehicle and object, which is named the distance between them. In addition, the 3DHMI data also records the moving speed and direction of each pedestrian, vehicle and object.

[0014] Furthermore, collecting the driver's braking habits includes the following sub-steps:

[0015] The vehicle itself is named the subject vehicle, and the pedestrians, vehicles, and objects around it are named as potential hazards. When analyzing any potential hazard, it is named as the target object.

[0016] Obtain the moving direction of the target object and the moving direction of the main vehicle, named the target direction and the main vehicle direction respectively. The target direction and the main vehicle direction are each a ray, and the endpoints of the ray are the target object and the main vehicle itself. Obtain the intersection of the target direction and the main vehicle direction, named the estimated collision point. The estimated collision point can change in real time.

[0017] The target's moving speed is labeled Vt, the vehicle's speed is labeled Ve, the distance between the vehicle and the estimated collision point is obtained, labeled D, and D / (Ve-Vt) is calculated. The result is named the estimated collision duration.

[0018] Each time the driver steps on the brake pedal, the minimum value of the different estimated collision durations is recorded, which is named the braking scenario duration. At the same time, the percentage of the driver's brake pedal stroke is recorded, which is named the braking stroke.

[0019] Analyze the driver's braking habits based on the braking scenario duration and braking distance.

[0020] Furthermore, analyzing the driver's braking habits based on the braking scenario duration and braking distance includes the following sub-steps:

[0021] A two-dimensional coordinate system is established with the braking scenario duration as the X-axis and the braking distance as the Y-axis. This is named the Braking Habit Analysis Chart. The braking distance is recorded into the Braking Habit Analysis Chart according to the braking scenario duration.

[0022] A discrete regression analysis is performed on the braking habit analysis graph, and a discrete regression function with the smallest standard deviation is selected and named as the braking characteristic function. The braking characteristic function is the braking habit.

[0023] Furthermore, customizing different risk limits for vehicles based on braking habits and environmental data includes the following sub-steps:

[0024] Analyze the duration of users' common braking scenarios based on their braking habits to obtain the common braking duration;

[0025] Different risk limits are customized for the vehicle based on the typical braking duration and environmental data.

[0026] Furthermore, analyzing the duration of the user's commonly used braking scenarios based on braking habits to obtain the commonly used braking duration includes the following sub-steps:

[0027] The coordinate points in the braking habit analysis graph are named braking habit analysis points, and cluster analysis is performed on the braking habit analysis points to obtain different trip clusters;

[0028] The number of braking habit analysis points in the trip cluster is counted and named as cluster data volume;

[0029] Obtain the three trip clusters with the largest amount of cluster data and name them common clusters. Calculate the average braking scenario duration in the common clusters and name it common reference duration.

[0030] The commonly used reference durations are named as the first risk duration, the second risk duration, and the third risk duration in order from large to small. The first risk duration, the second risk duration, and the third risk duration are commonly used braking durations.

[0031] Furthermore, customizing different risk limits for the vehicle based on the typical braking duration and environmental data includes the following sub-steps:

[0032] Obtain the safe reserve distance, represented by the symbol SRD, and obtain the maximum braking force of the vehicle, represented by the symbol MBF;

[0033] The first risk duration, the second risk duration, and the third risk duration are marked as T1, T2, and T3 respectively;

[0034] Substitute T1, T2, and T3 into the braking characteristic function to obtain the first, second, and third strokes, which are represented by symbols H1, H2, and H3 respectively;

[0035] By the formula BD1=SRD+Ve 2 / (2×H1×MBF×K) to calculate the first risk limit, where BD1 is the first risk limit and K is the road friction coefficient;

[0036] By the formula BD2=SRD+Ve 2 / (2×H2×MBF×K) to calculate the second risk limit, where BD2 is the second risk limit;

[0037] By the formula BD3=SRD+Ve 2 / (2×H3×MBF×K) to calculate the third risk limit, where BD3 is the third risk limit;

[0038] The first risk limit, the second risk limit and the third risk limit are risk limits.

[0039] Furthermore, based on the driving data, environmental data and risk limit, the collision risk of the vehicle is determined in real time to obtain the collision risk level, which includes the following sub-steps:

[0040] Obtain the estimated collision distances between the vehicle and all potential hazards, named the estimated collision distances, and obtain the minimum value of the estimated collision distances, named the minimum collision distance.

[0041] Determine the size of the minimum collision distance and the risk limit. If the minimum collision distance is less than or equal to the first risk limit, output the first collision risk signal; if the minimum collision distance is less than or equal to the second risk limit, output the second collision risk signal; if the minimum collision distance is less than or equal to the third risk limit, output the third collision risk signal, otherwise output the no collision risk signal.

[0042] Furthermore, providing a risk warning to the driver based on the collision accident risk level includes the following sub-steps:

[0043] If the first collision risk signal is output, the driver is advised to slow down through voice announcement;

[0044] If the second collision risk signal is output, the driver is advised to slow down through voice announcements, and a red halo is displayed on the central control screen and instrument panel;

[0045] If the third collision risk signal is output, the driver will be advised to slow down through voice broadcast, a red halo will be displayed on the central control screen and instrument panel, and a buzzer will be played to warn the driver.

[0046] Beneficial effects of the present invention: The present invention installs a multimodal on-board sensor in the vehicle and uses the multimodal on-board sensor to collect driving data and environmental data in real time during the vehicle's travel. The present invention then collects the driver's braking habits, and then analyzes the duration of the user's commonly used braking scenarios based on the braking habits to obtain the commonly used braking duration. The advantage of the present invention is that the commonly used braking duration of the driver can be adaptively found according to the driver's braking habits. The commonly used braking duration reflects the conditions under which the driver often brakes, so as to provide collision risk warnings in accordance with the driver's braking habits, thereby improving the applicability and effectiveness of vehicle collision risk warnings.

[0047] The present invention customizes different risk limits for vehicles through common braking duration and environmental data, and then judges the collision accident risk of the vehicle in real time based on the driving data, environmental data and risk limits to obtain the collision accident risk level. Finally, the driver is given a risk warning based on the collision accident risk level. The advantage is that different risk limits are customized for vehicles through common braking duration and environmental data, and the risk limits are dynamically adjusted according to the real-time vehicle speed and road friction coefficient to adapt to different road conditions. At the same time, sufficient braking space is reserved for the driver, which can effectively remind the driver to slow down in advance, rather than reminding the driver only in an emergency, thereby improving the accuracy and effectiveness of the vehicle collision risk warning. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 is a flow chart of the steps of the method of the present invention;

[0049] Figure 2 Schematic diagram of the body direction and target direction of the present invention;

[0050] Figure 3 This is a braking habit analysis diagram of the present invention;

[0051] Figure 4 Schematic diagram of trip clustering according to the present invention. DETAILED DESCRIPTION

[0052] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0053] Example 1, please refer to Figure 1 As shown, the present application provides a vehicle collision risk warning and accident warning method, comprising the following steps:

[0054] Step S1: installing a multimodal vehicle-mounted sensor in a vehicle; the multimodal vehicle-mounted sensor includes a laser radar, a high-definition camera, and a friction coefficient sensor;

[0055] In a specific implementation, the laser radar, high-definition camera and friction coefficient sensor are all existing multi-modal vehicle-mounted sensors and are not described in detail in this embodiment.

[0056] Step S2: Using multimodal on-board sensors to collect real-time driving data and environmental data during vehicle travel. The driving data is specifically 3D HMI data collected by a laser radar and a high-definition camera, and the environmental data is specifically the road friction coefficient detected by a friction coefficient sensor. The 3D HMI data is used to create and display 3D models of pedestrians, vehicles, and objects around the vehicle, while marking the distance between the vehicle and each pedestrian, vehicle, and object, named as the distance to each pedestrian, vehicle, and object. In addition, the 3D HMI data also records the speed and direction of each pedestrian, vehicle, and object.

[0057] In specific implementations, 3D HMI design can use three-dimensional graphics and animations to enable vehicles to more intuitively understand blind spots and obstacles around the vehicle, thereby improving the driving experience. The distance, moving speed and moving direction can also be obtained. This will not be further explained in detail in this embodiment.

[0058] Step S3, collecting the driver's braking habits; Step S3 includes the following sub-steps:

[0059] Step S301: The vehicle itself is named as the main vehicle, and pedestrians, vehicles, and objects around the vehicle are named as potential hazards. When analyzing any potential hazard, it is named as a target object.

[0060] See also Figure 2 As shown, in step S302, the moving direction of the target object and the moving direction of the main vehicle are obtained, which are named the target direction and the main vehicle direction respectively. The target direction and the main vehicle direction are each a ray, and the endpoints of the ray are the target object and the main vehicle itself. The intersection of the target direction and the main vehicle direction is obtained and named the estimated collision point. The estimated collision point can be changed in real time.

[0061] In specific implementation, Figure 2 For example, Figure 2 The figure shows the body direction and target direction of the vehicle and a target object, and the estimated collision point is obtained. Since the movement routes of the target object and the vehicle are not single but changeable, the body direction and target direction will change in real time, resulting in real-time changes in the estimated collision point. In addition, during the movement, the target direction and the body direction must be marked with reference to the lane. During the turn, the ray changes to a curve, but the property of the ray extending infinitely is retained to find the estimated collision point. Figure 2 In is the target direction and the body direction at the bend, Figure 2 In the figure, the dotted line represents the direction of the vehicle, and the potential line represents the direction of the target. This embodiment focuses on controlling the distance between the vehicle and the vehicle ahead during driving, ensuring that the vehicle ahead can respond in time when the vehicle ahead decelerates or brakes suddenly. Therefore, the estimated collision point is usually the location of the target itself.

[0062] Step S303: Label the target's moving speed Vt, label the vehicle's speed Ve, obtain the distance between the vehicle and the estimated collision point (D), calculate D / (Ve-Vt), and name the calculated result the estimated collision duration.

[0063] Step S304: Each time the driver steps on the brake pedal, the minimum value of the different estimated collision durations is recorded, which is named the braking scenario duration. The percentage of the driver's brake pedal stroke is also recorded, which is named the braking stroke.

[0064] In a specific implementation, the speed of the subject vehicle Ve is 60 km / h, approximately 16.67 m / s, and the target vehicle's speed is originally 70 km / h. There is no collision during normal driving, but the target vehicle suddenly brakes, causing Vt to drop to 40 km / h, approximately 11.11 m / s. At this time, D is 36 m, and the estimated collision time is calculated to be 6.47 s. The calculation result is rounded to two decimal places. The braking scene duration is the estimated collision time between the subject vehicle and the first vehicle in front, that is, 6.47 s. At this time, the driver's braking travel is recorded as 30%. The braking travel is a professional parameter and is not described in detail in this embodiment.

[0065] Step S305, analyzing the driver's braking habits based on the braking scene duration and braking distance;

[0066] Step S305 includes the following sub-steps:

[0067] See also Figure 3 As shown, in step S305.1, a two-dimensional coordinate system is established with the braking scene duration as the X-axis and the braking distance as the Y-axis, named the braking habit analysis chart, and the braking distance is recorded into the braking habit analysis chart according to the braking scene duration;

[0068] Step S305.2: Perform discrete regression analysis on the braking habit analysis graph, select the discrete regression function with the smallest standard deviation, and name it the braking characteristic function. The braking characteristic function is the braking habit

[0069] In the specific implementation, the braking habit analysis diagram is constructed as follows Figure 3 As shown, the braking characteristic function is obtained by analysis as Y=432.21X -1.336 , where Y is the braking distance and X is the braking scene duration.

[0070] Step S4, customizing different risk limits for the vehicle based on braking habits and environmental data; Step S4 includes the following sub-steps:

[0071] Step S401, analyzing the duration of the user's commonly used braking scenarios based on the user's braking habits to obtain the commonly used braking duration;

[0072] Step S401 includes the following sub-steps:

[0073] See also Figure 4 As shown, in step S401.1, the coordinate points in the braking habit analysis graph are named braking habit analysis points, and cluster analysis is performed on the braking habit analysis points to obtain different travel clusters;

[0074] Step S401.2: Count the number of braking habit analysis points in the trip cluster and name it cluster data volume;

[0075] Step S401.3: Obtain the three trip clusters with the largest cluster data volume, name them common clusters, and calculate the average braking scenario duration in the common clusters, name them common reference duration;

[0076] Step S401.4, naming the commonly used reference durations as the first risk duration, the second risk duration, and the third risk duration in descending order. The first risk duration, the second risk duration, and the third risk duration are commonly used braking durations;

[0077] In the specific implementation, the itinerary clusters are obtained through cluster analysis, such as Figure 4 As shown, they are named Cluster 1, Cluster 2, Cluster 3, Cluster 4 and Cluster 5 in order from left to right. It is easy to see that Cluster 1, Cluster 3 and Cluster 5 are the three trip clusters with the largest amount of cluster data, so they are named commonly used clusters. The average value of the X-axis value of the braking habit analysis point in the commonly used clusters is calculated to obtain the commonly used reference time. The commonly used reference time of each commonly used cluster is independent of each other. Through statistical calculation, the commonly used reference time of Cluster 1, Cluster 3 and Cluster 5 are 3.47s, 5.52s and 9.05s respectively. The calculation results are rounded to two decimal places. Through comparison, it is found that the first risk time is 9.05s, the second risk time is 5.52s, and the third risk time is 3.47s. The commonly used reference time means that the driver often brakes at these three collision times, which can better fit the driver's driving habits when customizing the risk threshold.

[0078] Step S402 , customizing different risk limits for the vehicle based on the typical braking duration and environmental data;

[0079] Step S402 includes the following sub-steps:

[0080] Step S402.1, obtaining a safety reserve distance, represented by the symbol SRD, and obtaining the maximum braking force of the vehicle, represented by the symbol MBF;

[0081] Step S402.2: Mark the first risk duration, the second risk duration, and the third risk duration as T1, T2, and T3, respectively;

[0082] Step S402.3: Substitute T1, T2, and T3 into the braking characteristic function to obtain the first, second, and third travels, which are represented by symbols H1, H2, and H3, respectively.

[0083] Step S402.4, by the formula BD1=SRD+Ve 2 / (2×H1×MBF×K) to calculate the first risk limit, where BD1 is the first risk limit and K is the road friction coefficient;

[0084] Step S402.5, by the formula BD2=SRD+Ve 2 / (2×H2×MBF×K) to calculate the second risk limit, where BD2 is the second risk limit;

[0085] Step S402.6, by the formula BD3=SRD+Ve 2 / (2×H3×MBF×K) to calculate the third risk limit, where BD3 is the third risk limit;

[0086] Step S402.7, the first risk limit, the second risk limit, and the third risk limit are the risk limits;

[0087] In specific implementation, the safety reserve distance is a preset value. Because any calculation has a certain error, it is necessary to ensure that the vehicle can still maintain a safe distance under the influence of the error, while also allowing the driver to have more reaction time. The safety reserve distance is set by the engineer. In this embodiment, the safety reserve distance SRD is set to 5m, and the maximum braking force MBF is obtained to be 8000N, T1 is 9.05s, T2 is 5.52s, and T3 is 3.47s. Substituting them into the calculation, H1, H2 and H3 are 22.8%, 44.1% and 82% respectively. The calculation result is rounded to one decimal place. The road friction coefficient K is detected to be 0.7, Ve is 60km / h, and the first risk limit BD1 is calculated to be 8.76m, the second risk limit BD2 is 6.94m, and the third risk limit BD3 is 6.05m.

[0088] Step S5: Determine the collision risk of the vehicle in real time based on the driving data, environmental data, and risk limit to obtain a collision risk level. Step S5 includes the following sub-steps:

[0089] Step S501: obtaining the estimated collision distances between the vehicle and all potential hazards, which is named the estimated collision distance, and obtaining the minimum value of the estimated collision distances, which is named the minimum collision distance;

[0090] Step S502: Determine the minimum collision distance and the risk limit. If the minimum collision distance is less than or equal to the first risk limit, output a first collision risk signal; if the minimum collision distance is less than or equal to the second risk limit, output a second collision risk signal; if the minimum collision distance is less than or equal to the third risk limit, output a third collision risk signal; otherwise, output a no collision risk signal.

[0091] In a specific implementation, the minimum collision distance is obtained to be 8.34m. By comparison, it is found that the minimum collision distance is less than the first risk limit, and a first collision risk signal is output;

[0092] Step S6: providing a risk warning to the driver based on the collision accident risk level. Step S6 includes the following sub-steps:

[0093] Step S601: If a first collision risk signal is output, the driver is advised to slow down through voice announcement;

[0094] Step S602: If the second collision risk signal is output, a voice announcement is made to advise the driver to slow down, and a red halo is displayed on the central control screen and instrument panel;

[0095] Step S603: If the third collision risk signal is output, the driver is advised to slow down through voice announcement, a red halo is displayed on the central control screen and instrument panel, and a buzzer is played to warn the driver;

[0096] In specific implementation, since the first collision risk signal is output, the driver is advised to slow down through voice broadcast; since the analysis is combined with the driver's braking habits, the collision risk warning for the driver can be more natural and will not bring more pressure to the driver. At the same time, the judgment criteria of the first collision risk signal, the second collision risk signal and the third collision risk signal are adaptively adjusted, which is more adaptable and accurate than the method of judging by fixed thresholds.

[0097] In Example 2, the present application provides an electronic device, which may include: a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other via the communication bus. The memory stores computer-readable instructions, and the processor can call the instructions in the memory. When the computer-readable instructions are executed by the processor, the steps in a vehicle collision risk warning and accident warning method are executed to achieve the following functions: installing a multimodal on-board sensor in the vehicle; using the multimodal on-board sensor to collect driving data and environmental data during the vehicle's driving process in real time; collecting the driver's braking habits; customizing different risk limits for the vehicle based on the braking habits and environmental data; judging the vehicle's collision accident risk in real time based on the driving data, environmental data, and risk limits to obtain a collision accident risk level; and providing a risk warning to the driver based on the collision accident risk level.

[0098] In addition, the logical instructions in the above-mentioned memory can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when sold or used as an independent product. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, and other media that can store program code.

[0099] Example 3. The present application also provides a computer program product, which includes a computer program stored on a computer-readable storage medium, and the computer program includes program instructions. When the program instructions are executed by a computer, the computer can execute a vehicle collision risk warning and accident warning method provided by the above methods, the method including: installing a multimodal vehicle-mounted sensor in a vehicle; collecting driving data and environmental data of the vehicle in real time during driving through the multimodal vehicle-mounted sensor; collecting the driver's braking habits; customizing different risk limits for the vehicle based on the braking habits and environmental data; judging the vehicle's collision accident risk in real time based on the driving data, environmental data and risk limits to obtain a collision accident risk level; and issuing a risk warning to the driver based on the collision accident risk level.

[0100] Example 4. The present application also provides a computer-readable storage medium. The present application provides a storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps in the above vehicle collision risk warning and accident warning method are executed to achieve the following functions: installing a multimodal vehicle-mounted sensor in the vehicle; collecting driving data and environmental data of the vehicle in real time through the multimodal vehicle-mounted sensor; collecting the driver's braking habits; customizing different risk limits for the vehicle based on the braking habits and environmental data; judging the vehicle's collision accident risk in real time based on the driving data, environmental data and risk limits to obtain a collision accident risk level; and issuing a risk warning to the driver based on the collision accident risk level.

[0101] Through the description of the above embodiments, the embodiments of the present invention can be provided as methods, systems, or computer program products. Based on this understanding, the essence of the above technical solutions or the portion that contributes to the prior art can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, or an optical disk, and includes a number of instructions for enabling a computer device (such as a personal computer, server, or network device) to execute the methods described in various embodiments or certain portions of the embodiments.

[0102] In the embodiments provided in this application, it should be understood that the disclosed system or method can be implemented in other ways. The embodiments described above are merely illustrative. For example, the division of modules or units is only a logical function division. There may be other division methods in actual implementation. For example, multiple modules or units can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interfaces, and the indirect coupling or communication connection of systems, modules and units can be electrical, mechanical or other forms.

[0103] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A vehicle collision risk warning and accident warning method, characterized in that: The steps include: Install multimodal onboard sensors in vehicles; The multimodal vehicle-mounted sensor collects driving data and environmental data of the vehicle in real time during driving; Collect drivers' braking habits; Customize different risk limits for vehicles based on braking habits and environmental data; Analyzing the duration of braking scenarios commonly used by users according to their braking habits to obtain commonly used braking durations, including: naming coordinate points in a braking habit analysis graph as braking habit analysis points, performing cluster analysis on the braking habit analysis points to obtain different travel clusters; counting the number of braking habit analysis points in the travel clusters, naming them as cluster data volume; obtaining three travel clusters with the largest cluster data volume, naming them as commonly used clusters, calculating the average value of the braking scenario durations in the commonly used clusters, naming them as commonly used reference durations; naming the commonly used reference durations as first risk duration, second risk duration, and third risk duration in order from largest to smallest, wherein the first risk duration, the second risk duration, and the third risk duration are the commonly used braking durations; Based on driving data, environmental data and risk limits, the collision risk of the vehicle is judged in real time to obtain the collision risk level; Provide risk warning to drivers based on the collision accident risk level.

2. A vehicle collision risk warning and accident warning method according to claim 1, characterized in that: A multimodal on-board sensor is installed in the vehicle, wherein the multimodal on-board sensor includes a laser radar, a high-definition camera, and a friction coefficient sensor.

3. A vehicle collision risk warning and accident warning method according to claim 2, characterized in that: The multimodal vehicle-mounted sensor collects driving data and environmental data in real time during the vehicle's driving process. The driving data is specifically 3DHMI data collected by a lidar and a high-definition camera, and the environmental data is specifically the road friction coefficient detected by a friction coefficient sensor. The 3DHMI data is used to perform 3D modeling and display of pedestrians, vehicles, and objects around the vehicle, and at the same time mark the distance between the vehicle and each pedestrian, vehicle, and object, named as the distance between them. In addition, the 3DHMI data also records the moving speed and direction of each pedestrian, vehicle, and object.

4. A vehicle collision risk warning and accident warning method according to claim 3, characterized in that: Collecting the driver's braking habits includes the following sub-steps: The vehicle itself is named the subject vehicle, and the pedestrians, vehicles, and objects around it are named as potential hazards. When analyzing any potential hazard, it is named as the target object. Obtain the moving direction of the target object and the moving direction of the main vehicle, named the target direction and the main vehicle direction respectively. The target direction and the main vehicle direction are each a ray, and the endpoints of the ray are the target object and the main vehicle itself. Obtain the intersection of the target direction and the main vehicle direction, named the estimated collision point. The estimated collision point can change in real time. The target's moving speed is labeled Vt, the vehicle's speed is labeled Ve, the distance between the vehicle and the estimated collision point is obtained, labeled D, and D / (Ve-Vt) is calculated. The result is named the estimated collision duration. Each time the driver steps on the brake pedal, the minimum value of the different estimated collision durations is recorded, which is named the braking scenario duration. At the same time, the percentage of the driver's brake pedal stroke is recorded, which is named the braking stroke. Analyze the driver's braking habits based on the braking scenario duration and braking distance.

5. The vehicle collision risk warning and accident warning method according to claim 4, characterized in that: Analyzing the driver's braking habits based on braking scenario duration and braking distance includes the following sub-steps: A two-dimensional coordinate system is established with the braking scenario duration as the X-axis and the braking distance as the Y-axis. This is named the Braking Habit Analysis Chart. The braking distance is recorded into the Braking Habit Analysis Chart according to the braking scenario duration. A discrete regression analysis is performed on the braking habit analysis graph, and a discrete regression function with the smallest standard deviation is selected and named as the braking characteristic function. The braking characteristic function is the braking habit.

6. The vehicle collision risk warning and accident warning method according to claim 5, characterized in that: Customizing different risk limits for vehicles based on braking habits and environmental data includes the following sub-steps: Different risk limits are customized for the vehicle based on the typical braking duration and environmental data.

7. The vehicle collision risk warning and accident warning method according to claim 6, characterized in that: Customizing different risk limits for vehicles based on typical braking duration and environmental data includes the following sub-steps: Obtain the safe reserve distance, represented by the symbol SRD, and obtain the maximum braking force of the vehicle, represented by the symbol MBF; The first risk duration, the second risk duration, and the third risk duration are marked as T1, T2, and T3 respectively; Substitute T1, T2, and T3 into the braking characteristic function to obtain the first, second, and third strokes, which are represented by symbols H1, H2, and H3 respectively; By the formula BD1=SRD+Ve 2 / (2×H1×MBF×K) to calculate the first risk limit, where BD1 is the first risk limit and K is the road friction coefficient; By the formula BD2=SRD+Ve 2 / (2×H2×MBF×K) to calculate the second risk limit, where BD2 is the second risk limit; By the formula BD3=SRD+Ve 2 / (2×H3×MBF×K) to calculate the third risk limit, where BD3 is the third risk limit; The first risk limit, the second risk limit and the third risk limit are risk limits.

8. The vehicle collision risk warning and accident warning method according to claim 7, characterized in that: The collision risk of a vehicle is determined in real time based on driving data, environmental data, and risk limits. Obtaining the collision risk level includes the following sub-steps: Obtain the estimated collision distances between the vehicle and all potential hazards, named the estimated collision distances, and obtain the minimum value of the estimated collision distances, named the minimum collision distance. Determine the size of the minimum collision distance and the risk limit. If the minimum collision distance is less than or equal to the first risk limit, output the first collision risk signal; if the minimum collision distance is less than or equal to the second risk limit, output the second collision risk signal; if the minimum collision distance is less than or equal to the third risk limit, output the third collision risk signal, otherwise output the no collision risk signal.

9. The vehicle collision risk warning and accident warning method according to claim 8, characterized in that: Providing risk warning to drivers based on the collision risk level includes the following sub-steps: If the first collision risk signal is output, the driver is advised to slow down through voice announcement; If the second collision risk signal is output, the driver is advised to slow down through voice announcements, and a red halo is displayed on the central control screen and instrument panel; If the third collision risk signal is output, the driver will be advised to slow down through voice broadcast, a red halo will be displayed on the central control screen and instrument panel, and a buzzer will be played to warn the driver.

Images