TWI823613B - Anti-collision warning method, vehicle-mounted device, and storage medium - Google Patents

Abstract

The present application provides an anti-collision warning method, a vehicle-mounted device, and a storage medium. The anti-collision warning method includes: fusing acquired radar information and image information; identifying an obstacle in a direction of travel of a vehicle according to the fused radar information and image information; determining motion parameters of the obstacle and the vehicle according to the radar information and the image information; calculating collision time between the vehicle and the obstacle according to the motion parameters, and issuing an anti-collision warning. This application can assist the user in driving safely.

Description

Anti-collision warning method, vehicle-mounted device and storage medium

The present invention relates to the technical field of safe driving, and in particular, to an anti-collision warning method, a vehicle-mounted device and a storage medium.

The vehicle's collision avoidance warning system can help users avoid collisions and other traffic accidents by continuously detecting road conditions in the direction of the vehicle's travel. However, existing anti-collision warning systems have problems such as inaccurate recognition of obstacle categories and inaccurate warning time.

In view of the above, it is necessary to provide an anti-collision warning method, vehicle-mounted device and storage medium that can effectively improve the accuracy of obstacle recognition and the accuracy of warning time, and assist users in safe driving.

The anti-collision warning method includes: fusing the acquired radar information and image information; identifying obstacles in the direction of vehicle travel based on the fused radar information and image information; and determining the obstacles based on the radar information and the image information. The motion parameters of the obstacle and the vehicle are calculated; the collision time between the vehicle and the obstacle is calculated based on the motion parameters, and an anti-collision warning is issued.

Optionally, determining the motion parameters of the obstacle and the vehicle based on the radar information and the image information includes: determining the relative distance and relative speed of the obstacle and the vehicle based on the radar information; The predicted traveling distance of the obstacle is predicted based on the image information and the relative speed.

Optionally, calculating the impact time between the vehicle and the obstacle based on the movement parameters and issuing an anti-collision warning includes: calculating a first impact countdown based on the movement parameters; calculating a second impact based on the movement parameters. Countdown; if the first impact countdown is less than the second impact countdown, issue the anti-collision warning.

Optionally, the method further includes: jointly calibrating the radar device that acquires the radar information and the camera device that acquires the image information; and combining the point cloud included in the radar information and the image included in the image information. Performing a fusion that includes projecting the point cloud to the image.

Optionally, identifying obstacles in the direction of vehicle travel based on the fused radar information and image information includes: identifying the target object in the image based on a preset deep neural network, and determining the surrounding area of the target object. box; classify the fused point cloud based on the bounding box of the target object, and cluster the classified point cloud; obtain a bounding box based on the clustered point cloud and the bounding box of the target object; based on the bounding box The box determines whether the target object is the obstacle and the type of obstacle.

Optionally, predicting the predicted forward distance of the obstacle based on the image information and the relative speed includes: obtaining the interval time between taking two adjacent images, and the corresponding distance of the obstacle in the two images. Two relative velocities; calculate the acceleration of the obstacle based on the interval time and the two relative velocities; calculate the predicted forward distance based on the interval time and the acceleration.

Optionally, calculating the first impact countdown based on the motion parameters includes: making the first impact countdown proportional to the relative distance between the obstacle and the vehicle and the predicted forward distance of the obstacle, and Inversely proportional to the relative speed of the obstacle and the vehicle.

Optionally, calculating the second impact countdown based on the motion parameters includes: calculating the braking distance of the vehicle based on the vehicle's speed and gravity acceleration, wherein the braking distance is proportional to the square of the vehicle speed. , and is inversely proportional to the acceleration of gravity; let the second impact countdown be directly proportional to the braking distance, and inversely proportional to the relative speed.

The computer-readable storage medium stores at least one instruction. When the at least one instruction is executed by a processor, the anti-collision warning method or the anti-collision warning method is implemented.

The vehicle-mounted device includes a storage and at least one processor. At least one instruction is stored in the storage. When the at least one instruction is executed by the at least one processor, the collision avoidance warning method is implemented.

Compared with the prior art, the anti-collision warning method provided by the embodiments of the present application determines the motion parameters of the obstacle and the vehicle based on radar information and image information, and calculates the collision time between the vehicle and the obstacle based on the motion parameters, which can effectively improve the detection of obstacles. The accuracy of recognition and the accuracy of warning time assist users in driving safely.

3: Vehicle-mounted device

30: Anti-collision warning system

31:Storage

32: Processor

33:Radar device

34:Camera device

S1~S4: steps

In order to more clearly explain the technical solutions in the embodiments of the present application or the conventional technology, the following will briefly introduce the drawings needed to describe the embodiments or the conventional technology. Obviously, the drawings in the following description are only This is an embodiment of the present application. For those of ordinary skill in the art, other drawings can be obtained based on the provided drawings without exerting creative efforts.

Figure 1 is a flow chart of the anti-collision warning method provided by an embodiment of the present application.

FIG. 2 is an architectural diagram of a vehicle-mounted device provided by an embodiment of the present application.

In order to more clearly understand the above objects, features and advantages of the present application, the present application will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that, as long as there is no conflict, the embodiments of the present application and the features in the embodiments can be combined with each other.

Many specific details are set forth in the following description to facilitate a full understanding of the present application. The described embodiments are only some, rather than all, of the embodiments of the present application. Based on the embodiments in this application, those of ordinary skill in the art can obtain the results without any creative efforts. All other embodiments obtained fall within the scope of protection of this application.

Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the technical field to which this application belongs. The terminology used herein in the description of the application is for the purpose of describing specific embodiments only and is not intended to limit the application.

In one embodiment, the vehicle's anti-collision warning system can quickly and effectively detect potential dangerous situations by continuously detecting road conditions in the direction of travel of the vehicle, and provide different sound and/or visual reminders to help users avoid rear-end collisions, Unintentional deviation from the lane, collision with pedestrians and other obstacles and other traffic accidents. The existing anti-collision warning system has problems such as inaccurate recognition of obstacle categories and inaccurate warning time.

In order to solve the above problems, the anti-collision warning method provided by the embodiment of the present application is based on the fusion of monocular vision with millimeter wave radar, determines the motion parameters of obstacles and vehicles based on radar information and image information, and calculates the motion parameters of vehicles and obstacles based on the motion parameters. The collision time can effectively improve the accuracy of obstacle recognition and warning time, and assist users in safe driving.

Refer to FIG. 1 , which is a flow chart of a collision avoidance warning method according to a preferred embodiment of the present application.

In this embodiment, the anti-collision warning method can be applied to a vehicle-mounted device installed in a vehicle (such as the vehicle-mounted device shown in Figure 2). For vehicle-mounted devices that require anti-collision warning, the vehicle-mounted device can be directly installed on the vehicle-mounted device. The anti-collision warning function provided by the method of the embodiment of the present application can be integrated or run on the vehicle-mounted device in the form of a software development kit (Software Development Kit, SDK).

As shown in Figure 1, the anti-collision warning method specifically includes the following steps. According to different needs, the order of the steps in the flow chart can be changed, and some steps can be omitted.

Step S1: Fusion of the acquired radar information and image information.

In one embodiment, the vehicle-mounted device may include a plurality of sensor devices, such as a ranging sensor and an image sensor. Specifically, the ranging sensor may include a radar device (such as a millimeter wave radar). , the image sensor may include a camera device (such as a monocular camera device). In the embodiment of the present application, the vehicle-mounted device can be a device configured in the vehicle, installed in the vehicle and having a corresponding software system to execute various instructions; it can also be an external device connected to the vehicle through communication to realize the control of vehicle data. acquisition and control of the vehicle.

The radar device can be installed in the vehicle, such as on the front windshield of the vehicle, to obtain point clouds (for example, three-dimensional point clouds) in the direction of travel of the vehicle; the camera device can be installed on the vehicle's front windshield. Positions such as the front windshield are used to obtain images of the vehicle's direction of travel (for example, two-dimensional images). The camera device may be a driving recorder installed in the vehicle, or it may be one or more cameras. The cameras may be installed on the vehicle, or they may be independent devices connected to the vehicle through a network or other means. Practical applications are not limited to the above examples.

In one embodiment, the method further includes: jointly calibrating a radar device that acquires the radar information and a camera device that acquires the image information; and combining the point cloud included in the radar information with the point cloud included in the image information. The images are fused, and the fusion includes projecting the point cloud onto the image.

In the multi-sensor detection system composed of the radar device and the camera device, the two sensors need to be jointly calibrated before fusing the radar information and the image information, so as to obtain the points and points of the point cloud. The corresponding relationship between the pixels of the image projects the three-dimensional point cloud into the two-dimensional image to complete the fusion of radar information and image information.

In one embodiment, the basic principle of jointly calibrating and fusing radar information and image information includes: obtaining external parameters (rotation matrix, translation vector, etc.) of the radar device and camera device respectively, and obtaining the location where the radar device is located based on the external parameters. The transformation matrix between the world coordinate system and the coordinate system where the camera device is located (for example, the coordinate transformation relationship matrix calculated based on the Perspective-n-Point algorithm); based on the transformation matrix, the point in the three-dimensional point cloud coordinate system is converted into Project to the three-dimensional coordinate system where the camera device is located; obtain the internal parameters of the camera device (focal length, principal point, tilt coefficient, distortion coefficient, etc.) by performing camera calibration on the camera device, and eliminate the distortion effect of the convex lens of the camera device based on the internal parameters, A point projected into the three-dimensional coordinate system in which the camera device is located is projected into the two-dimensional image. Specifically, a variety of calibration tools can be used to implement the above process, for example, APOLLO's sensor equipment calibration tool, AUTOWARE's CalibrationTookit module, etc.

In one embodiment, a multi-sensor detection system can effectively improve the vehicle's environmental awareness and safety performance. Among them, the integration of ranging sensors and image sensors has obvious advantages in obtaining environmental information and target recognition. Among them, millimeter-wave radar has the advantages of wide detection range, less affected by weather, etc. compared with other ranging sensors, and has good adaptability.

In other embodiments, the radar device and the camera device can also be installed at other locations on the vehicle, for example, at the rear windshield of the vehicle, so as to issue an anti-collision warning in conjunction with subsequent steps when the vehicle is about to be rear-ended.

Step S2: Identify obstacles in the direction of vehicle travel based on the fused radar information and image information.

In one embodiment, identifying obstacles in the direction of vehicle travel based on the fused radar information and image information includes: identifying the target object in the image based on a preset deep neural network, and determining the target object bounding box; classify the fused point cloud based on the bounding box of the target object, and cluster the classified point cloud; obtain a bounding box based on the clustered point cloud and the bounding box of the target object; based on the The bounding box determines whether the target object is the obstacle and the type of the obstacle.

In one embodiment, the deep neural network may be YOLOv3 based on YOLOv3 (You Only Look Once). YOLOv3 uses a fully convolutional network to divide the image into multiple sub-regions and predict the boundaries of the object based on the multiple sub-regions. The probability of the object category that the frame and each sub-region belongs to is removed through the non-maximum suppression algorithm to remove redundant bounding boxes, thereby distinguishing the objects in the image and the category of each object, and determining the bounding box (two-dimensional bounding box) of the target object. frame). It can save calculation time and improve the detection accuracy of object recognition.

In one embodiment, the target objects may include pedestrians, vehicles, guardrails, lane lines, waste paper, etc.

In one embodiment, after determining the bounding box of the target object, the method is based on the fused Based on the radar information and image information, identifying obstacles in the direction of vehicle travel specifically includes: based on the two-dimensional bounding box of the target object in the two-dimensional image, classifying the points in the three-dimensional point cloud projected onto the two-dimensional image; The clustering algorithm removes noise points from the classified 3D point cloud and points that are far away from each type of 3D point cloud (for example, screening far away points through a preset distance threshold); the clustered 3D points are The two-dimensional bounding box image of the cloud and the target object is input into a deep learning network (such as a deep convolutional neural network), and the three-dimensional bounding box of the clustered three-dimensional point cloud is obtained by regression, thereby obtaining the target of the corresponding bounding box. The type of object, as well as the location and type of obstacles in the target object (for example, non-low or flat target objects such as pedestrians, vehicles, guardrails, etc.).

In one embodiment, based on the two-dimensional bounding box and the three-dimensional bounding box, two category detections in two dimensions and three dimensions are performed on the fused point cloud and the target object in the image, which can effectively improve the object quality. Category detection accuracy.

Step S3: Determine motion parameters of the obstacle and the vehicle based on the radar information and the image information.

In one embodiment, determining the motion parameters of the obstacle and the vehicle based on the radar information and the image information includes: determining the relative distance and relative distance between the obstacle and the vehicle based on the radar information. Speed; predict the predicted forward distance of the obstacle based on the image information and the relative speed.

In one embodiment, the radar device can calculate the relative speed of the obstacle and the radar device based on the Doppler Effect principle. Specifically, the Doppler effect principle includes: when an obstacle approaches the radar antenna, the frequency of the reflected signal will be higher than the frequency of the transmitter; conversely, when the obstacle moves away from the antenna, the frequency of the reflected signal will be lower than the frequency of the transmitter. Therefore, the relative speed can be calculated based on the change in frequency (when an obstacle moves toward the radar, the Doppler frequency is positive; when the obstacle moves away from the radar, the Doppler frequency is negative).

In one embodiment, the three-dimensional point cloud collected by the radar device includes depth information of each point's distance from the radar device, and the minimum depth value among all points of the obstacle can be used as the point cloud depth of the obstacle. In addition, when installing the radar device, the distance between the radar device and the vehicle body can be determined. (for example, the distance between the radar device installed at the front windshield of the vehicle and the front of the vehicle), thereby calculating the relative distance between the obstacle and the vehicle according to the preset formula, where the preset formula is: relative distance = obstacle The point cloud depth of the object - the distance between the radar device and the vehicle body.

In one embodiment, predicting the predicted forward distance of the obstacle based on the image information and the relative speed includes: obtaining the interval time between taking two adjacent images, and the obstacles in the two images. Corresponding two relative velocities; calculate the acceleration of the obstacle based on the interval time and the two relative velocities; calculate the predicted forward distance based on the interval time and the acceleration.

In one embodiment, the interval time t can be obtained based on the frame rate of images captured by the camera device. For example, when the frame rate is 30 frames per second, the interval time t=1/30 seconds. In addition, the obstacle can be assumed to be an object that moves in a straight line with uniform acceleration. The relative speed with the first time among the two relative speeds is recorded as V ₁ , and the relative speed with the later time among the two relative speeds is recorded as V 1 . is V ₂ , then the acceleration a of the obstacle can be calculated according to the formula of uniform acceleration V ₂ =V ₁ +at (values include positive numbers, negative numbers, and 0).

In one embodiment, when the anti-collision warning method provided by the embodiment of the present application calculates the predicted forward distance, the following situations include but are not limited to: Situation 1: The obstacle and the vehicle are traveling in the same direction and the acceleration of the obstacle is Negative value; Case 2: The obstacle and the vehicle are moving relative to each other and the acceleration of the obstacle is positive; Case 1: The obstacle and the vehicle are traveling in the same direction and the acceleration of the obstacle is negative; Case 3: The obstacle is a stationary object .

For example, the first situation can be that the vehicle is driving in one lane of a double lane, and the vehicle in front of it that is traveling in the same direction is slowing down; the second situation can be that the vehicle is driving in a single lane, and that the vehicle in front that is driving in the opposite direction is slowing down. The vehicle is braking and moving slowly; the third situation may be that the vehicle is driving and obstacles such as roadblocks appear directly in front. Specifically, through the fused radar information and image information, the specific situation of the vehicle can be easily determined, and no detailed description will be given.

In one embodiment, the following formula can be used to calculate the predicted forward distance S in cases one and two: S=v ₀ t+at ² /2, where v ₀ represents the last picture obtained by the anti-collision warning system. The speed of the corresponding obstacle; in case three, the predicted forward distance S=0. Among them, it is assumed that the anti-collision warning system in the embodiment of the present application can determine whether to give a warning within the interval time t. In other embodiments, the time t in this formula can also be set according to the actual calculated speed of the collision avoidance warning system.

Step S4: Calculate the collision time between the vehicle and the obstacle according to the motion parameters, and issue an anti-collision warning.

In one embodiment, the impact time includes a countdown time (for example, 1 second) when the vehicle collides with the obstacle. Calculating the impact time between the vehicle and the obstacle according to the motion parameters and issuing an anti-collision warning includes: calculating a first impact countdown according to the motion parameters; calculating a second impact countdown according to the motion parameters; if If the first impact countdown is less than the second impact countdown, the anti-collision warning is issued.

In one embodiment, the first impact countdown can be calculated based on the motion parameters for each of the above situations. For example, in case 1, the calculation of the first impact countdown based on the motion parameters includes: making the first impact countdown It is directly proportional to the relative distance between the obstacle and the vehicle, the predicted forward distance of the obstacle, and inversely proportional to the relative speed of the obstacle and the vehicle. For example, let the first impact countdown = (relative distance + predicted forward distance) / relative speed.

In other embodiments, such as situation 2, calculating the first impact countdown based on the motion parameters includes: making the first impact countdown proportional to the relative distance between the obstacle and the vehicle, and making the first impact countdown proportional to the relative distance between the obstacle and the vehicle, and The obstacle is inversely proportional to the relative speed of the vehicle and the predicted forward distance of the obstacle. For example, let the first impact countdown = (relative distance - predicted forward distance) / relative speed.

In one embodiment, calculating the second impact countdown based on the motion parameters includes: calculating the braking distance of the vehicle based on the vehicle's speed and gravity acceleration, where the The braking distance is proportional to the square of the vehicle speed and inversely proportional to the acceleration due to gravity; the second impact countdown is proportional to the braking distance and inversely proportional to the relative speed. For example, let the braking distance = (self-vehicle speed * self-vehicle speed) / (2 * 0.9 * g), and let the second impact countdown = braking distance / relative speed, where g represents the acceleration of gravity.

In one embodiment, the anti-collision warning can be output in a preset manner, which may include but is not limited to displaying text or image content, outputting audio, vibration, etc.

In one embodiment, the anti-collision warning method provided by this application is based on the fusion of monocular vision with millimeter wave radar, determines the motion parameters of obstacles and vehicles based on radar information and image information, and calculates the distance between vehicles and obstacles based on the motion parameters. The first impact countdown and the second impact countdown. When the first impact countdown is less than the second impact countdown, an anti-collision warning is issued, which can effectively improve the accuracy of obstacle recognition and the accuracy of the warning time, and assist users in safe driving.

The above-mentioned Figure 1 introduces the anti-collision early warning method of the present application in detail. The functional modules of the software system that implements the anti-collision early warning method and the hardware device architecture that implements the anti-collision early warning method are introduced below with reference to Figure 2 .

It should be understood that the above embodiments are for illustration only, and the scope of the patent application is not limited by this structure.

Refer to FIG. 2 , which is a schematic structural diagram of a vehicle-mounted device according to a preferred embodiment of the present application.

In the preferred embodiment of the present application, the vehicle-mounted device 3 includes a memory 31 , at least one processor 32 , at least one radar device 33 , and at least one camera device 34 . Those skilled in the art should understand that the structure of the vehicle-mounted device shown in Figure 2 does not constitute a limitation of the embodiment of the present application. It can be a bus-type structure or a star-shaped structure. The vehicle-mounted device 3 can also include a The illustrations illustrate more or less additional hardware or software, or different arrangements of components.

In some embodiments, the vehicle-mounted device 3 includes a terminal that can automatically perform numerical calculations and/or information processing according to preset or stored instructions. Its hardware includes but is not limited to microprocessors, special integrated circuits, Programmable logic gate arrays, digital signal processors and embedded Built-in equipment, etc.

It should be noted that the vehicle-mounted device 3 is only an example. If other existing or future electronic products can be adapted to this application, they should also be included in the protection scope of this application and are included here by reference.

In some embodiments, the storage 31 is used to store program codes and various data. For example, the memory 31 can be used to store the anti-collision warning system 30 installed in the vehicle-mounted device 3 , and realize high-speed and automatic access to programs or data during the operation of the vehicle-mounted device 3 . The storage 31 includes read-only memory (Read-Only Memory, ROM), programmable read-only memory (Programmable Read-Only Memory, PROM), erasable programmable read-only memory (Erasable Programmable Read). -Only Memory (EPROM), One-time Programmable Read-Only Memory (OTPROM), Electronically Erasable Programmable Read-Only Memory (EEPROM) , Compact Disc Read-Only Memory (CD-ROM) or other optical disk storage, magnetic disk storage, tape storage, or any other computer-readable storage medium that can be used to carry or store data.

In some embodiments, the at least one processor 32 may be composed of an integrated circuit, for example, it may be composed of a single packaged integrated circuit, or it may be composed of multiple integrated circuits packaged with the same function or different functions. , including one or more central processing units (CPUs), microprocessors, digital signal processing chips, graphics processors and a combination of various control chips. The at least one processor 32 is the control core (Control Unit) of the vehicle-mounted device 3. It uses various interfaces and lines to connect various components of the entire vehicle-mounted device 3, and runs or executes programs stored in the memory 31 or module, and calls the data stored in the memory 31 to perform various functions of the vehicle-mounted device 3 and process data, such as performing the anti-collision warning function shown in Figure 1 .

In some embodiments, the collision avoidance warning system 30 runs in the vehicle-mounted device 3 . The anti-collision warning system 30 may include a plurality of functional modules composed of program code segments. Said prevention The program codes of each program segment in the collision warning system 30 can be stored in the memory 31 of the vehicle-mounted device 3 and executed by at least one processor 32 to implement the anti-collision warning function shown in FIG. 1 .

In this embodiment, the anti-collision warning system 30 can be divided into multiple functional modules according to the functions it performs. The module referred to in this application refers to a series of computer program segments that can be executed by at least one processor and can complete fixed functions, which are stored in the memory.

Although not shown, the vehicle-mounted device 3 may also include a power supply (such as a battery) that supplies power to various components. Preferably, the power supply may be logically connected to the at least one processor 32 through a power management device, so that the power supply can be implemented through the power management device. Manage functions such as charging, discharging, and power consumption management. The power supply may also include one or more DC or AC power supplies, recharging devices, power failure test circuits, power converters or inverters, power status indicators and other arbitrary components. The vehicle-mounted device 3 may also include a variety of sensors, Bluetooth modules, Wi-Fi modules, etc., which will not be described again here.

It should be understood that the above embodiments are for illustration only, and the scope of the patent application is not limited by this structure.

The above-mentioned integrated unit implemented in the form of a software function module can be stored in a computer-readable storage medium. The above-mentioned software function module is stored in a storage medium and includes a number of instructions to cause a vehicle-mounted device (which can be a server, a personal computer, etc.) or a processor to execute part of the method described in each embodiment of the present application.

Program codes are stored in the memory 31 , and the at least one processor 32 can call the program codes stored in the memory 31 to perform related functions. The program code stored in the memory 31 can be executed by the at least one processor 32 to implement the functions of each module to achieve the purpose of anti-collision warning.

In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only illustrative. For example, the division of the modules is only a logical function division, and the actual implementation There can be other division methods.

The modules described as separate components may or may not be physically separated. The components shown as modules may or may not be physical units, that is, they may be located in one place, or they may be distributed to multiple networks. on the unit. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

In addition, each functional module in various embodiments of the present application can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of hardware plus software function modules.

It is obvious to those skilled in the art that the present application is not limited to the details of the above-described exemplary embodiments, and that the present application can be implemented in other specific forms without departing from the spirit or essential characteristics of the present application. Therefore, the embodiments should be regarded as illustrative and non-restrictive from any point of view, and the scope of the present application is defined by the appended claims rather than the above description, and it is therefore intended that those falling within the claims All changes within the meaning and scope of the equivalent elements are included in this application. Any reference designation in a request shall not be construed as limiting the request to which it refers. Furthermore, it is obvious that the word "including" does not exclude other elements or the singular does not exclude the plural. Multiple units or devices stated in the device request may also be implemented by one unit or device through software or hardware. Words such as first and second are used to indicate names and do not indicate any specific order.

Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application and are not limiting. Although the present application has been described in detail with reference to the above preferred embodiments, those of ordinary skill in the art will understand that the technical solutions of the present application can be modified. The technical solution may be modified or equivalently substituted without departing from the spirit and scope of the technical solution of the present application.

S1~S4: steps

Claims (9)

An anti-collision warning method, applied to vehicle-mounted devices, wherein the method includes: fusing the acquired radar information and image information; identifying obstacles in the direction of the vehicle's travel based on the fused radar information and image information; according to the The radar information and the image information determine the motion parameters of the obstacle and the vehicle, including: determining the relative distance and relative speed of the obstacle and the vehicle based on the radar information; based on the image information and The relative speed predicts the predicted forward distance of the obstacle; the collision time between the vehicle and the obstacle is calculated based on the motion parameters, and an anti-collision warning is issued. The anti-collision warning method according to claim 1, wherein calculating the collision time between the vehicle and the obstacle based on the motion parameters and issuing the anti-collision warning includes: calculating a first impact countdown based on the motion parameters ; Calculate the second impact countdown based on the motion parameters; if the first impact countdown is less than the second impact countdown, issue the anti-collision warning. The anti-collision early warning method according to claim 1, wherein the method further includes: jointly calibrating the radar device for acquiring the radar information and the camera device for acquiring the image information; The cloud is fused with the image included in the image information, and the fusion includes projecting the point cloud onto the image. The anti-collision warning method as described in claim 3, wherein identifying obstacles in the direction of vehicle travel based on the fused radar information and image information includes: identifying obstacles in the image based on a preset deep neural network target object, and determine the bounding box of the target object; classify the fused point cloud based on the bounding box of the target object, and cluster the classified point cloud; according to the clustered point cloud and the bounding box of the target object Bounding box: Obtain a bounding box; determine whether the target object is the obstacle and the type of the obstacle based on the bounding box Don't. The anti-collision warning method according to claim 1, wherein predicting the predicted forward distance of the obstacle based on the image information and the relative speed includes: obtaining the interval time between taking two adjacent images, and Two relative velocities corresponding to the obstacles in the two images; calculate the acceleration of the obstacle based on the interval time and the two relative velocities; calculate the predicted advance based on the interval time and the acceleration distance. The anti-collision warning method according to claim 2, wherein calculating the first impact countdown based on the motion parameters includes: combining the first impact countdown with the relative distance between the obstacle and the vehicle, the The predicted distance traveled by an obstacle is directly proportional to the relative speed of the obstacle and the vehicle. The anti-collision warning method of claim 2, wherein calculating the second impact countdown based on the motion parameters includes: calculating the braking distance of the vehicle based on the vehicle's speed and gravity acceleration, wherein the braking distance The stopping distance is proportional to the square of the vehicle speed and inversely proportional to the acceleration of gravity; let the second impact countdown be proportional to the stopping distance and inversely proportional to the relative speed. A computer-readable storage medium, wherein the computer-readable storage medium stores at least one instruction. When the at least one instruction is executed by a processor, the anti-collision warning method as described in any one of claims 1 to 7 is implemented. . A vehicle-mounted device, wherein the vehicle-mounted device includes a storage and at least one processor, at least one instruction is stored in the storage, and when the at least one instruction is executed by the at least one processor, it implements claims 1 to 7 The anti-collision warning method described in any one of the above.

Images