Vehicle side collision control method, device, electronic device and storage medium

Abstract

The present application provides a vehicle side collision control method, device, electronic device and storage medium, the method comprising: obtaining the lateral distance of an obstacle and current motion information of the vehicle, determining the collision control time according to the current motion information and the lateral distance of the obstacle, the collision control time being used to characterize the time corresponding to at least one of avoiding a collision and causing a collision between the vehicle and obstacles on both sides, and performing at least one of lateral collision warning control and lateral collision braking control on the vehicle based on the collision control time, wherein the lateral distance of the obstacle is obtained by inputting the images of the two sides of the vehicle body to be inspected corresponding to when the vehicle starts to turn into the environmental model on both sides of the vehicle body for obstacle detection, and the environmental model on both sides of the vehicle body is obtained by training based on historical images on both sides of the vehicle body and multiple obstacle label information; the present application reduces or even avoids side scraping collisions through intelligent perception of obstacles on both sides of the vehicle body, protects the safety of the vehicle body, and improves the safety of vulnerable traffic participants in the environment on both sides of the vehicle body.

Description

Technical Field

[0001] The present application relates to the field of vehicle safety technology, and in particular to a vehicle side collision control method, device, electronic device and storage medium. Background Art

[0002] With the development of automobile intelligence and software-defined cars, the degree of automobile intelligence has been greatly improved. The Around View Monitor (AVM) has become a standard feature of new energy vehicles or mid-to-high-end fuel vehicles. It uses an ultra-wide-angle fisheye lens to capture images, and uses a special algorithm to correct the distortion and stitch the captured images to form a panoramic image system around the object. In combination with ultrasonic radar, it monitors the distance information to the surrounding environment in real time, providing the driver with a wider field of view and more reminders when reversing into a garage or parking sideways. However, AVM will only transmit the environmental information around the vehicle body to the screen for display and sound and light reminders, and will not interfere with vehicle control. Whether there is a scratch depends on the driver's decision. In addition, the environmental information provided by AVM is more of a collision reminder for the rear, left front, right front, left rear and right rear of the vehicle. It is difficult to monitor and remind the left and right sides of the vehicle, especially the scratches behind the middle A-pillar on both sides. However, during the low-speed turning and parking of the car, the rear door and rear fender of the vehicle are the areas with the most serious scratches. Due to the driver's driving experience, the driver's concentration, external environmental interference, driving or parking area restrictions, etc., there is a greater risk of scratches on the part behind the A-pillars on both sides of the vehicle.

[0003] The current electronic rearview mirror uses high-definition cameras and ultra-clear screens to obtain the environment on both sides of the car body. It can adaptively adjust the field of view for scenarios such as high speed, turning and reversing, but it will only display the environment on both sides of the car body on the screen, obtaining a larger and clearer field of view. The driver still needs to continuously observe the image on the monitor during driving to avoid scratches on both sides of the car body.

[0004] Common scenarios where the sides of the car body are scratched include: 1. When the vehicle enters or leaves the garage, there are vehicles parked on both sides of the target parking space, the target parking space is narrow or the distance in front of the target parking space is short. The driver needs to pay attention to multiple perspectives at the same time when parking the vehicle in / out of the parking space; 2. When driving in a garage or a narrow road with many curves. In this scenario, the road width is limited and most of them have sharp turns. During the turning process, the distance to obstacles on both sides of the car body is not accurately controlled, and the timing and angle of the turn are wrong, which can easily lead to the car body being scratched against the surrounding obstacles; 3. The road posts and stone piers placed on the edge of the road are relatively short in shape and are very common on internal roads. They are used to divide road driving areas and prohibit passage. This scenario usually occurs in scenarios such as internal roads in residential areas, business districts and industrial parks. In this scenario, the road environment is relatively open, and the driver will pay more attention to the front of the vehicle, thereby ignoring the sides of the vehicle body. In addition, the road posts and stone piers are relatively short and easy to collide.

[0005] Electronic rearview mirrors and panoramic monitoring imaging systems are currently the main means of perceiving the environment around the vehicle body, but they still have the following problems: electronic rearview mirrors and AVM only provide field of view and obstacle warnings at the four corners of the vehicle body, resulting in a limited warning area, and unable to provide prompts for areas with collision risks in the field of view on both sides of the vehicle body; electronic rearview mirrors only transmit images of the field of view on both sides to the monitor and are unable to perceive the surrounding field of view. AVM provides obstacle warnings for the four corners and rear of the vehicle body, but lacks perception of obstacles on both sides of the vehicle body, resulting in a lack of intelligent perception; the vehicle cannot be intelligently controlled according to obstacles on both sides of the vehicle body to effectively avoid collisions between the two sides of the vehicle body; in common side-scratching scenarios, the driver needs to pay attention to multiple directions of the vehicle body at the same time, and the field of view of the electronic exterior rearview mirror cannot be transmitted to the driver in real time, and there is a lack of real-time response; the driver can only obtain information on both sides of the vehicle body by actively observing the screen, and cannot be reminded of obstacles with scratch risks and lacks active interaction. Summary of the invention

[0006] The present application provides a vehicle side collision control method, device, electronic device and storage medium to solve the above-mentioned technical problems that it is difficult to intelligently perceive obstacles on both sides of the vehicle body through electronic rearview mirrors and panoramic monitoring imaging systems, and it is difficult to achieve real-time response and intelligent control based on obstacles on both sides of the vehicle body.

[0007] In one embodiment of the present application, the present application provides a vehicle side collision control method, including: obtaining the vehicle's obstacle lateral distance and current motion information, the obstacle lateral distance is used to characterize the lateral distance between the obstacles on both sides and the vehicle body when it is detected that the vehicle starts to turn; determining the collision control time according to the current motion information and the obstacle lateral distance, the collision control time is used to characterize the time corresponding to at least one of avoiding collision and causing collision between the vehicle and the obstacles on both sides; performing at least one of lateral collision warning control and lateral collision braking control on the vehicle based on the collision control time; wherein, the obstacle lateral distance is obtained by inputting the images of the vehicle body to be inspected corresponding to the time when the vehicle starts to turn into the vehicle body side environment model for obstacle detection, and the vehicle body side environment model is trained based on historical vehicle body side images and multiple obstacle label information.

[0008] In one embodiment of the present application, the collision control time is determined according to the current motion information and the lateral distance of the obstacle, including: determining the maximum steering angle for the vehicle to avoid collision with the obstacles on both sides according to the current steering angle, the current vehicle speed and the lateral distance of the obstacle; and / or, respectively decomposing the current vehicle speed and the current acceleration in the direction of the vehicle wheelbase to obtain the lateral vehicle speed and the lateral acceleration, and determining the collision occurrence time of redundant collision between the vehicle and the obstacles on both sides based on the difference between the lateral vehicle speed, the lateral acceleration, the lateral distance of the obstacle and a preset safety distance; wherein the current motion information includes the current steering angle, the current vehicle speed and the current acceleration, and the collision control time includes the collision occurrence time and the steering control time corresponding to the maximum steering angle.

[0009] In one embodiment of the present application, at least one of lateral collision warning control and lateral collision braking control is performed on the vehicle based on the collision control time, including: determining the vehicle control time according to a preset control time threshold and the collision control time, the vehicle control time including at least one of the vehicle warning time and the vehicle braking time; comparing the collision occurrence time and the vehicle control time to obtain a first comparison result, or comparing the current turning time and the vehicle control time to obtain a second comparison result, the current motion information including the current turning time; performing at least one of lateral collision warning control and lateral collision braking control on the vehicle according to the first comparison result, or performing at least one of lateral collision warning control and lateral collision braking control on the vehicle according to the second comparison result.

[0010] In one embodiment of the present application, the vehicle control time is determined according to a preset control time threshold and the collision control time, including: if the collision occurrence time exists in the collision control time, the collision occurrence time is used as the collision braking time, and the collision warning time is determined according to the preset collision time threshold and the collision occurrence time; if the steering control time exists in the collision control time, the steering control time is used as the steering braking time, and at least one of the steering warning times is determined according to the preset steering time threshold and the steering control time; if both the collision braking time and the steering braking time exist, the minimum braking time value of the collision braking time and the steering braking time is determined as the vehicle braking time; if both the collision warning time and the steering warning time exist, the minimum warning time value of the collision warning time and the steering warning time is determined as the vehicle warning time.

[0011] In one embodiment of the present application, the method for determining the lateral distance of the obstacle includes: if it is detected that the vehicle starts to turn, the images of both sides of the vehicle body to be inspected when the vehicle starts to turn are collected through the external electronic rearview mirror; the images of both sides of the vehicle body to be inspected are input into the environmental model of both sides of the vehicle body for obstacle detection, and the current lateral distance between the obstacles on both sides and the vehicle body is obtained; the current lateral distance closest to the vehicle body is used as the lateral distance of the obstacle.

[0012] In one embodiment of the present application, before obtaining the lateral distance of the obstacle of the vehicle, the method also includes: obtaining historical images of both sides of the vehicle body and multiple obstacle label information, wherein the historical images of both sides of the vehicle body are obtained by an image acquisition device in an exterior electronic rearview mirror; performing obstacle detection on an area of ​​interest in the historical images of both sides of the vehicle body according to each obstacle label information to obtain actual coordinates of the obstacle, wherein the area of ​​interest is used to characterize an image area where there is a risk of collision; performing depth estimation on the image corresponding to the area of ​​interest based on a preset device internal reference of the image acquisition device to obtain a depth image, establishing a three-dimensional grid map corresponding to the depth image, and obtaining the three-dimensional coordinates of the obstacle in a preset image acquisition coordinate system according to the three-dimensional grid map and the actual coordinates of the obstacle; converting the three-dimensional coordinates of the obstacle to a preset vehicle body coordinate system, and obtaining the historical lateral distance between the obstacles on both sides and the vehicle body according to the converted three-dimensional coordinates of the obstacle; wherein the vehicle body side environment model includes the obstacle detection, the establishment of the three-dimensional grid map, and the coordinate system conversion training.

[0013] In one embodiment of the present application, obstacle detection is performed on the region of interest in the historical vehicle body side images according to each obstacle label information to obtain the two-dimensional coordinates of the obstacle, including: performing image preprocessing on the historical vehicle body side images to obtain a first vehicle body side image, the image preprocessing includes color restoration of the non-vehicle area in the historical vehicle body side images by average exposure; dividing the region of interest in the first vehicle body side images, scaling the image corresponding to the region of interest according to a preset zoom factor to obtain a second vehicle body side image, and performing pixel filling on the second vehicle body side image to obtain a third vehicle body side image with the same length and width, wherein the region of interest is located in the first region; performing obstacle detection on the third vehicle body side image according to each obstacle label information to obtain target detection coordinates, and restoring the target detection coordinates to the first vehicle body side image according to the preset zoom factor to obtain the actual coordinates of the obstacle; wherein the first vehicle body side image includes the first region located in the middle of the image, the second region and the third region located on both sides of the image, the second region is used to characterize the region where the vehicle body is located, and the third region is used to characterize the region whose distance difference with the second region is greater than or equal to the preset distance threshold.

[0014] In one embodiment of the present application, the present application provides a vehicle side collision control device, including: an acquisition module, used to acquire the vehicle's obstacle lateral distance and current motion information, the obstacle lateral distance is used to characterize the lateral distance between the obstacles on both sides and the vehicle body when it is detected that the vehicle starts to turn; a time determination module, used to determine the collision control time according to the current motion information and the obstacle lateral distance, the collision control time is used to characterize the time corresponding to at least one of avoiding collision and causing collision between the vehicle and the obstacles on both sides; a control module, used to perform at least one of lateral collision warning control and lateral collision braking control on the vehicle based on the collision control time; wherein the obstacle lateral distance is obtained by inputting the images of the vehicle body to be inspected corresponding to the time when the vehicle starts to turn into the vehicle body side environment model for obstacle detection, and the vehicle body side environment model is trained based on historical vehicle body side images and multiple obstacle label information.

[0015] In one embodiment of the present application, the present application provides an electronic device, which includes: one or more processors; a storage device for storing one or more programs, and when the one or more programs are executed by the one or more processors, the electronic device implements the vehicle side collision control method as described in any one of the above embodiments.

[0016] In one embodiment of the present application, the present application provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor of a computer, the computer executes the vehicle side collision control method described in any one of the above embodiments.

[0017] Beneficial effects of the embodiments of the present application: The present application provides a vehicle side collision control method, device, electronic device and storage medium. The embodiments of the present application detect obstacles by inputting the images of both sides of the vehicle body to be inspected corresponding to when the vehicle starts to turn into the environment model on both sides of the vehicle body to obtain the lateral distance of the obstacle, thereby realizing intelligent perception of obstacles on both sides of the vehicle body; and performing at least one of lateral collision warning control and lateral collision braking control on the vehicle according to the collision control time determined by the current motion information and the lateral distance of the obstacle, thereby realizing the function of real-time response and intelligent control according to the obstacles on both sides of the vehicle body, which can effectively avoid scratches and collisions on both sides of the vehicle body caused by the driver's immature driving skills or lack of concentration, protect the safety of the vehicle body, and improve the safety of two-wheeled vehicles and pedestrians in the environment on both sides of the vehicle body.

[0018] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The drawings herein are incorporated into the specification and constitute a part of the specification, showing embodiments consistent with the present application, and together with the specification, are used to explain the principles of the present application. Obviously, the drawings described below are only some embodiments of the present application, and for those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0020] Figure 1 A schematic diagram showing an exemplary system architecture to which the technical solution of the embodiments of the present application can be applied;

[0021] Figure 2 A schematic flow chart of a vehicle side collision control method according to an embodiment of the present application is shown;

[0022] Figure 3 A schematic diagram of establishing a three-dimensional model of an obstacle according to an embodiment of the present application is shown;

[0023] Figure 4 A schematic diagram showing the division of an area of ​​interest according to an embodiment of the present application is shown;

[0024] Figure 5 A schematic diagram of depth estimation according to an embodiment of the present application is shown;

[0025] Figure 6A schematic diagram of an implementation of a vehicle side collision control method according to an embodiment of the present application is shown;

[0026] Figure 7 A block diagram of a vehicle side collision control device according to an embodiment of the present application is shown;

[0027] Figure 8 A schematic diagram of the structure of a computer system suitable for implementing an electronic device of an embodiment of the present application is shown. DETAILED DESCRIPTION

[0028] The following describes the embodiments of the present application through specific examples, and those skilled in the art can easily understand other advantages and effects of the present application from the contents disclosed in this specification. The present application can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present application. It should be noted that the following embodiments and features in the embodiments can be combined with each other without conflict.

[0029] It should be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present application, and thus the drawings only show components related to the present application rather than being drawn according to the number, shape and size of components in actual implementation. In actual implementation, the type, quantity and proportion of each component may be changed at will, and the component layout may also be more complicated.

[0030] In the following description, a large number of details are discussed to provide a more thorough explanation of the embodiments of the present application. However, it is obvious to those skilled in the art that the embodiments of the present application can be implemented without these specific details. In other embodiments, well-known structures and devices are shown in the form of block diagrams rather than in detail to avoid making the embodiments of the present application difficult to understand.

[0031] See also Figure 1 , Figure 1 A schematic diagram of an exemplary system architecture to which the technical solution of the embodiment of the present application can be applied is shown. Figure 1 As shown, the system architecture may include an image acquisition device 101 and a vehicle 102. The image acquisition device 101 is located on a bracket of an external electronic rearview mirror, and is used to acquire images of both sides of the vehicle body to be inspected when the vehicle starts to turn, and transmit the images to the vehicle 102, so as to obtain the lateral distance of the obstacle through the images of both sides of the vehicle body to be inspected.

[0032] Exemplarily, after obtaining the lateral distance of the obstacle and the current motion information of the vehicle 102, the collision control time is determined according to the current motion information and the lateral distance of the obstacle, and the collision control time is used to characterize the time corresponding to at least one of avoiding a collision and causing a collision between the vehicle and the obstacles on both sides; at least one of lateral collision warning control and lateral collision braking control is performed on the vehicle based on the collision control time; wherein, the lateral distance of the obstacle is used to characterize the lateral distance between the obstacles on both sides and the vehicle body when it is detected that the vehicle starts to turn, and the lateral distance of the obstacle is obtained by inputting the images of the two sides of the vehicle body to be inspected corresponding to the time when the vehicle starts to turn into the environmental model on both sides of the vehicle body for obstacle detection, and the environmental model on both sides of the vehicle body is trained based on historical images on both sides of the vehicle body and multiple obstacle label information.

[0033] In the related technologies, electronic rearview mirrors and panoramic monitoring imaging systems are currently the main means of perceiving the surrounding environment of a vehicle body and are mainly used for information display. It is difficult to intelligently perceive obstacles on both sides of the vehicle body through electronic rearview mirrors and panoramic monitoring imaging systems, and it is difficult to achieve technical problems such as real-time response and intelligent control based on obstacles on both sides of the vehicle body.

[0034] In order to solve the above technical problems, the present application provides a vehicle side collision control method, device, electronic device and storage medium. The implementation details of the technical solution of the embodiment of the present application are elaborated in detail below.

[0035] See also Figure 2 , Figure 2 FIG. 2 shows a flow chart of a vehicle side collision control method according to an embodiment of the present application. Figure 2 As shown, in an exemplary embodiment, the vehicle side collision control method includes at least steps S210 to S230, which are described in detail as follows:

[0036] Step S210, obtaining the lateral distance of the vehicle to the obstacle and the current motion information.

[0037] Among them, the lateral distance of the obstacle is used to characterize the lateral distance between the obstacles on both sides and the vehicle body when the vehicle starts to turn. The lateral distance of the obstacle is obtained by inputting the images of the two sides of the vehicle body to be inspected corresponding to when the vehicle starts to turn into the vehicle body side environment model for obstacle detection. The vehicle body side environment model is trained based on historical vehicle body side images and multiple obstacle label information.

[0038] In one embodiment of the present application, the current motion information is acquired in real time.

[0039] In one embodiment of the present application, a method for determining the lateral distance of an obstacle includes: if it is detected that the vehicle starts to turn, images of both sides of the vehicle body to be inspected when the vehicle starts to turn are collected through the external electronic rearview mirror; the images of both sides of the vehicle body to be inspected are input into the environmental model of both sides of the vehicle body for obstacle detection, and the current lateral distance between the obstacles on both sides and the vehicle body is obtained; and the current lateral distance closest to the vehicle body is used as the lateral distance of the obstacle.

[0040] In one embodiment of the present application, the start of vehicle turning is detected by monitoring the steering wheel angle signal, thereby performing vehicle side collision control.

[0041] In one embodiment of the present application, if the current speed is lower than the preset control speed and the steering wheel angle signal is detected, the current vehicle body side images are collected as the vehicle body side images to be inspected. Through the vehicle body side environment model established in the early stage, the obstacle lateral distance of the obstacles on both sides from the vehicle body can be obtained.

[0042] In one embodiment of the present application, before obtaining the lateral distance of the obstacle of the vehicle, the method also includes: obtaining historical images of both sides of the vehicle body and multiple obstacle label information, the historical images of both sides of the vehicle body are obtained by an image acquisition device in an exterior electronic rearview mirror; performing obstacle detection on an area of ​​interest in the historical images of both sides of the vehicle body according to each obstacle label information to obtain the actual coordinates of the obstacle, the area of ​​interest is used to characterize the image area where there is a risk of collision; performing depth estimation on the image corresponding to the area of ​​interest based on a preset device internal parameter of the image acquisition device to obtain a depth image, establishing a three-dimensional grid map corresponding to the depth image, and obtaining the three-dimensional coordinates of the obstacle in a preset image acquisition coordinate system according to the three-dimensional grid map and the actual coordinates of the obstacle; converting the three-dimensional coordinates of the obstacle to a preset vehicle body coordinate system, and obtaining the historical lateral distance between the obstacles on both sides and the vehicle body according to the converted three-dimensional coordinates of the obstacle; wherein the vehicle body side environment model includes obstacle detection, establishment of a three-dimensional grid map, and coordinate system conversion training.

[0043] In one embodiment of the present application, obstacle detection is performed on an area of ​​interest in a historical image on both sides of a vehicle body according to information of each obstacle tag to obtain two-dimensional coordinates of the obstacle, including: performing image preprocessing on the historical image on both sides of the vehicle body to obtain a first image on both sides of the vehicle body, the image preprocessing includes color restoration of a non-vehicle area in the historical image on both sides of the vehicle body by average exposure; dividing the area of ​​interest in the first image on both sides of the vehicle body, scaling the image corresponding to the area of ​​interest according to a preset zoom factor to obtain a second image on both sides of the vehicle body, and performing pixel filling on the second image on both sides of the vehicle body to obtain a third image on both sides of the vehicle body with the same length and width, wherein the area of ​​interest is located in the first area; performing obstacle detection on the third image on both sides of the vehicle body according to information of each obstacle tag to obtain target detection coordinates, and restoring the target detection coordinates to the first image on both sides of the vehicle body according to the preset zoom factor to obtain actual coordinates of the obstacle; wherein the first image on both sides of the vehicle body includes a first area located in the middle of the image, a second area and a third area located on both sides of the image, the second area is used to characterize an area where the vehicle body is located, and the third area is used to characterize an area whose distance difference from the second area is greater than or equal to a preset distance threshold.

[0044] In one embodiment of the present application, the present application can perform vehicle side collision control for a vehicle traveling at a low speed. For example, the speed of the low speed travel includes 0 to 30 km / h.

[0045] In one embodiment of the present application, the present application can also perform vehicle side collision control for vehicles above 30 km / h.

[0046] In one embodiment of the present application, the environment on both sides of the vehicle body is collected and modeled by the image acquisition device of the Class III field of view of the exterior electronic rearview mirror to obtain the lateral distance of the obstacle, and the time to occur or avoid a collision is calculated in real time based on the lateral distance of the obstacle, so as to perform scratch warning and emergency braking to avoid the vehicle body from scratching or colliding with vulnerable traffic participants such as stationary pedestrians and stationary two-wheeled vehicles.

[0047] In one embodiment of the present application, the preset installation position of the exterior electronic rearview mirror needs to meet the basic field of view requirements, and must include the entire vertical downward and rearward position of the front door handles on both sides of the vehicle body. The image acquisition frame rate needs to be greater than 60FPS, the resolution is 1920*1280, and the image is not distorted.

[0048] In one embodiment of the present application, in the process of arranging the image acquisition device in the exterior electronic rearview mirror, it is arranged to the maximum extent strictly in accordance with the driver's habit of viewing a traditional rearview mirror. In the horizontal direction of the field of view of the image acquisition device, the front door handle can be seen close to the side of the vehicle body, and in the vertical direction, the rear wheel contact point can be seen. The installation angle of the image acquisition device is calibrated according to the above conditions.

[0049] In one embodiment of the present application, historical images of both sides of the vehicle body are acquired by an arranged image acquisition device.

[0050] In one embodiment of the present application, the image acquisition device includes a camera.

[0051] In one embodiment of the present application, image preprocessing includes at least one of image format conversion, color restoration, distortion removal, denoising, and enhancement. The historical images on both sides of the vehicle body collected by the camera have not been processed by image signal processing (ISP), and the image format is still raw image data (RAW), which is inconsistent with the image format actually used for training. The historical images on both sides of the vehicle body must first be restored to RGB888 format, that is, restored to 24-bit true color. Due to the installation position of the camera, 1 / 5 of the image area in the historical images on both sides of the vehicle body is always the vehicle body area. When the camera is exposed, the light and scene of this part do not change much. Therefore, the vehicle body area is eliminated during the exposure process, and the color of the remaining non-vehicle area is restored by average exposure. A polynomial distortion removal algorithm is used to correct the radial distortion and tangential distortion caused by the lens. After correction, a Gaussian filtering algorithm is used to remove salt and pepper noise. In the enhancement stage, adaptive equalization is used to enhance the visual contrast of the image.

[0052] In one embodiment of the present application, after image preprocessing, the image data resolution of the first vehicle body images on both sides is 1920*1280. The present application performs obstacle detection on the region of interest (ROI) in the first vehicle body images on both sides to reduce the power consumption, memory, computing power and cost of the entire system and ensure the real-time performance of the system.

[0053] In one embodiment of the present application, see Figure 3 , Figure 3 FIG. 2 shows a schematic diagram of establishing a three-dimensional obstacle model according to an embodiment of the present application. Figure 3 As shown in the figure, the common scratch scenes on both sides of the car body are identified through the target detection algorithm and the label information of each obstacle, and the region of interest is divided to obtain the ROI, so as to realize the environmental perception from coarse to fine; by performing depth estimation on the image corresponding to the region of interest, the depth information of each pixel in the image corresponding to the region of interest is estimated; a depth image is established according to the depth information of each pixel; the depth image is further converted into a three-dimensional grid map; the environmental map of both sides of the car body is obtained through the three-dimensional grid map to predict the position of obstacles.

[0054] In one embodiment of the present application, see Figure 4 , Figure 4 FIG. 2 shows a schematic diagram of dividing the region of interest according to an embodiment of the present application. Figure 4 As shown, the camera arrangement position is combined for calibration, the image area of ​​the first vehicle body on both sides is divided, and the left one-fifth and the right one-fifth are used as the second area where the vehicle body is located or the third area far from the vehicle body to indicate that there is no risk of scratching. Figure 4 The solid line area and the bold solid line area are marked in the figure; the remaining middle area is used as the preliminary region of interest, and the image size of the region of interest is 1152*1280. Figure 4 The dotted area indicates that the region of interest is still very large. If depth estimation is performed directly, the computing power required is high, and the cost and real-time performance cannot be guaranteed.

[0055] In one embodiment of the present application, in order to ensure real-time performance, the image corresponding to the region of interest is scaled according to the high side: the 1280 image is scaled to 320, the preset scaling factor is 0.25, and the width is scaled at the same ratio, and the resolution of the second vehicle body side image is 288*320. Then, 16 pixels are filled in the upper and lower sides of the second vehicle body side image to obtain a 320*320 third vehicle body side image. Common scratch scene detection is performed on the third vehicle body side image, and the initial detection coordinates (x, y, w, h) category and confidence of the scratch scene obstacle are output. The initial detection coordinates whose confidence meets the preset probability threshold are used as the target detection coordinates, and the target detection coordinates are restored to the actual first vehicle body side image according to the scaling factor, and the region of interest is further clarified according to the actual coordinates of the obstacle.

[0056] In one embodiment of the present application, depth estimation is performed on the image corresponding to the region of interest. Figure 5 , Figure 5 FIG. 2 shows a schematic diagram of depth estimation according to an embodiment of the present application. Figure 5The depth estimation shown in the figure adopts the absolute depth estimation algorithm (ZeroDepth) based on deep learning. Compared with the traditional depth estimation algorithm, the algorithm solves the two problems of absolute scale and generalization at the same time. By converting the input in any format into a two-dimensional embedding (2D embedding), cross-domain generality is achieved, and then the corresponding 2D embedding is mapped to the potential representation, and the camera intrinsic parameter is added to implicitly infer the physical properties such as the size and shape of the object, so as to achieve generalization of different data sets and camera parameters. The camera intrinsic parameter (K) in the preset device intrinsic parameter and the image (Image) corresponding to the region of interest are converted into a camera embedding vector (Cameraembedding) and an image embedding vector (Image embedding) respectively. During the conversion process of the image embedding vector, the resolution will gradually decrease, and the dimension of the feature map will be further improved, so as to extract deep features. In order to improve the depth estimation model for the difference in appearance and geometric fields, the camera embedding vector and the image embedding vector are fused and encoded (Encoder), and then the camera embedding vector is fused again after encoding for decoding (Decoder). The depth information of each pixel is further predicted through decoding to obtain a depth image (Depth image).

[0057] In one embodiment of the present application, in order to further reduce the amount of calculation and facilitate the later collision warning, the obtained depth image is processed to establish a three-dimensional grid map with depth. The three-dimensional environment on both sides of the vehicle body is established based on the two-dimensional image, and the motion trajectory of the two sides of the vehicle body is predicted based on the three-dimensional grid map, so as to realize the scratch warning and braking on both sides. The three-dimensional environment around the vehicle body now established is based on the preset image acquisition coordinate system, which is quite different from the actual preset vehicle body coordinate system.

[0058] In one embodiment of the present application, the camera is mounted on the bracket of the external electronic rearview mirror, with the optical center of the camera as the origin, the optical axis as the z-axis, and the x and y axes respectively parallel to the x and y directions of the image, to establish a preset image acquisition coordinate system.

[0059] In one embodiment of the present application, a preset vehicle body coordinate system is established with the center of the rear axle of the vehicle as the origin, the vehicle body wheelbase direction as the first axis, the vehicle body length direction as the second axis, and the vehicle body height direction as the third axis.

[0060] In one embodiment of the present application, the first axis includes an x-axis, the second axis includes a y-axis, and the third axis includes a z-axis.

[0061] In one embodiment of the present application, a preset vehicle body coordinate system is established with the center of the rear axle of the vehicle as the origin, the x-axis points to the right along the carrier axis, the y-axis points forward, and the z-axis, x-axis, and y-axis point to the sky in accordance with the right-hand coordinate rule.

[0062] In one embodiment of the present application, the three-dimensional scene established based on the preset image acquisition coordinate system is transformed into the preset vehicle body coordinate system by obtaining the preset relative coordinates between the vehicle body and the camera and the preset transformation matrix in advance.

[0063] In one embodiment of the present application, for a three-dimensional environment established under a preset vehicle body coordinate system, obstacles on both sides close to the vehicle body are expanded with an expansion distance of 5 cm, and the expanded three-dimensional grid map is used to analyze the vehicle body collision risk, thereby obtaining the historical lateral distance between the obstacles on both sides and the vehicle body.

[0064] In one embodiment of the present application, the historical lateral distance can be obtained through the converted three-dimensional coordinates of the obstacle and half the width of the vehicle body.

[0065] Step S220, determining a collision control time according to the current motion information and the lateral distance of the obstacle, wherein the collision control time is used to represent the time corresponding to at least one of avoiding a collision and causing a collision between the vehicle and the obstacles on both sides.

[0066] In one embodiment of the present application, the collision control time is determined based on the current motion information and the lateral distance of the obstacle, including: determining the maximum steering angle for the vehicle to avoid collision with obstacles on both sides based on the current steering angle, the current vehicle speed and the lateral distance of the obstacle; and / or, respectively decomposing the current vehicle speed and the current acceleration in the direction of the vehicle wheelbase to obtain the lateral vehicle speed and the lateral acceleration, and determining the collision occurrence time of redundant collision between the vehicle and the obstacles on both sides based on the difference between the lateral vehicle speed, the lateral acceleration, the lateral distance of the obstacle and the preset safety distance; wherein the current motion information includes the current steering angle, the current vehicle speed and the current acceleration, and the collision control time includes the collision occurrence time and the steering control time corresponding to the maximum steering angle.

[0067] In one embodiment of the present application, a collision avoidance model based on the turning point position is used to take the position where the vehicle starts turning as the turning point position according to the lateral distance of the obstacles on both sides from the vehicle body, freeze the lateral distance of the obstacles at the turning point position, and combine the current steering angular velocity determined by the current vehicle speed and the current steering angle to obtain a collision model established by the current steering angular velocity and the vehicle steering angle.

[0068] In one embodiment of the present application, after sensing an obstacle, the position where the steering begins is used as the turning point, and the lateral distance of the obstacle is calculated through the images of both sides of the vehicle body to be inspected at the turning point, so as to predict the vehicle's driving trajectory in real time in advance. At the current vehicle speed, the maximum steering angle of the vehicle's steering wheel without a collision is calculated in real time as the braking steering angle for emergency braking.

[0069] In one embodiment of the present application, a redundant collision model based on the collision occurrence time is as follows: the current speed of the vehicle is recorded as v (m / s), the current acceleration is recorded as a (m / s 2 ), the preset safety distance relative to the nearest obstacles on both sides is d(m), and the actual distance is s(m). The above distance is the distance between each point on both sides of the vehicle body and the obstacles on both sides in the x-axis direction of the preset vehicle body coordinate system, that is, the actual distance is the lateral distance of the obstacle.

[0070] In one embodiment of the present application, the present application is mainly aimed at stationary targets on both sides, so it is assumed that the speed and acceleration of objects with collision risk on both sides of the vehicle body are zero. Because the collision occurs on both sides of the vehicle body, the current speed and current acceleration need to be decomposed into the x-axis and y-axis along the vehicle body coordinate system according to the preset kinematic model to obtain the lateral vehicle speed v corresponding to the x-axis. x and lateral acceleration a x , and the longitudinal speed v corresponding to the y-axis y and longitudinal acceleration a y The collision is mainly considered in the x-axis direction, and the formula of the redundant collision model is shown in formula (1):

[0071]

[0072] Among them, s is the actual distance, d is the preset safety distance, and v x is the lateral speed, t is the time of collision, a x is the lateral acceleration.

[0073] In one embodiment of the present application, the key to formula (1) is the time-to-collision (TTC). According to formula (1), the time-to-collision of a collision can be solved as follows:

[0074]

[0075] Among them, t is the time when the collision occurs, v x is the lateral speed, s is the actual distance, d is the preset safety distance, a x is the lateral acceleration.

[0076] In one embodiment of the present application, since the vehicle body is in motion and the complexity of vehicle body steering is relatively high, the present application will adopt a lateral collision avoidance model based on the redundancy of the turning point position and collision time at the start of steering.

[0077] Step S230: performing at least one of lateral collision warning control and lateral collision braking control on the vehicle based on the collision control time.

[0078] In one embodiment of the present application, the side collision includes a scratch scenario.

[0079] In one embodiment of the present application, at least one of lateral collision warning control and lateral collision braking control is performed on the vehicle based on the collision control time, including: determining the vehicle control time according to a preset control time threshold and the collision control time, the vehicle control time including at least one of the vehicle warning time and the vehicle braking time; comparing the collision occurrence time and the vehicle control time to obtain a first comparison result, or comparing the current turning time and the vehicle control time to obtain a second comparison result, the current motion information including the current turning time; performing at least one of lateral collision warning control and lateral collision braking control on the vehicle according to the first comparison result, or performing at least one of lateral collision warning control and lateral collision braking control on the vehicle according to the second comparison result.

[0080] In one embodiment of the present application, the vehicle control time is determined according to a preset control time threshold and a collision control time, including: if the collision occurrence time exists in the collision control time, the collision occurrence time is used as the collision braking time, and the collision warning time is determined according to the preset collision time threshold and the collision occurrence time; if the collision control time exists in the collision control time, the steering control time is used as the steering braking time, and at least one of the steering warning times is determined according to the preset steering time threshold and the steering control time; if both the collision braking time and the steering braking time exist, the minimum braking time value of the collision braking time and the steering braking time is determined as the vehicle braking time; if both the collision warning time and the steering warning time exist, the minimum warning time value of the collision warning time and the steering warning time is determined as the vehicle warning time.

[0081] In one embodiment of the present application, the time corresponding to the maximum steering angle is denoted as T a-b , which is the steering braking time.

[0082] In one embodiment of the present application, the preset turning time threshold includes a preset percentage threshold.

[0083] In one embodiment of the present application, a preset percentage threshold of the maximum steering angle is used as the warning time, denoted as T a-w , that is, the turn warning time. The preset percentage threshold may be 80%, which is only an example and does not impose any limitation on the preset percentage threshold of the present application.

[0084] In one embodiment of the present application, the preset collision time threshold includes a control percentage or a control safety time. The control percentage of the collision occurrence time, or the difference between the collision occurrence time and the control safety time is used as the collision warning time.

[0085] In one embodiment of the present application, the collision braking time is recorded as T t-b , the collision warning time is recorded as T t-w .

[0086] In one embodiment of the present application, in view of the limited sensing distance and position during the turning process, some scenes at corners cannot be directly detected, and obstacle information can only be obtained after turning. In order to improve the stability and reliability of the system, the present application adopts a multi-model redundant design, using the minimum time of trigger control as the measurement indicator of model redundancy, thereby obtaining the vehicle warning time T w and vehicle braking time T b .

[0087] In one embodiment of the present application, the vehicle warning time is determined as shown in formula (3):

[0088] T w =min(T t-w ,T a-w ) Formula (3)

[0089] Among them, T w is the vehicle warning time, T t-w is the collision warning time, T a-w It is the turn warning time.

[0090] In one embodiment of the present application, the vehicle braking time is determined as shown in formula (4):

[0091] T b =min(T t-b ,T a-b ) Formula (4)

[0092] Among them, T b is the vehicle braking time, T t-b is the collision braking time, T a-b is the steering braking time.

[0093] In one embodiment of the present application, the collision time and the vehicle control time are compared to obtain a first comparison result, including: if the collision time is less than the vehicle braking time, the first comparison result is determined as requiring vehicle emergency braking; if the collision time is less than the vehicle warning time, the first comparison result is determined as requiring vehicle collision warning.

[0094] In one embodiment of the present application, the vehicle control time is determined according to a preset control time threshold and a collision control time, and also includes: if only a collision warning time exists, the collision warning time is determined as the vehicle warning time; if only a steering warning time exists, the steering warning time is determined as the vehicle warning time; if only a collision braking time exists, the collision braking time is determined as the vehicle braking time; if only a steering braking time exists, the steering braking time is determined as the vehicle braking time.

[0095] In one embodiment of the present application, the current turning time and the vehicle control time are compared to obtain a second comparison result including: if the absolute value of the difference between the current turning time and the vehicle braking time is less than or equal to a preset first threshold, the second comparison result is determined as requiring control of vehicle emergency braking; if the absolute value of the difference between the current turning time and the vehicle warning time is less than or equal to a preset second threshold, the second comparison result is determined as requiring control of a vehicle collision warning.

[0096] In one embodiment of the present application, when t <T w When t <T b When the vehicle is in emergency, the brake system is controlled.

[0097] In one embodiment of the present application, the present application supports the operation functions of vehicle side collision control such as closing, opening, and level setting. Opening / closing can be set through the vehicle computer, and can be opened or closed according to the needs of the driver. Closing will not affect the basic functions of the exterior electronic rearview mirror; in order to meet the preferences and needs of different users, a total of three preset emergency braking gears, namely comfort, normal, and aggressive, are set for adjustment. Under the three preset emergency braking gears, due to the differences in the preset safety distance, the preset control threshold, and the preset percentage threshold, the vehicle warning time T w and vehicle braking time T b Different, resulting in different inertia forces of braking parameters.

[0098] In one embodiment of the present application, see Figure 6 , Figure 6 FIG. 1 shows a schematic diagram of an implementation of a vehicle side collision control method according to an embodiment of the present application. Figure 6As shown, the electronic exterior rearview mirror camera: collects images of both sides of the vehicle body to be detected when the vehicle starts to turn through a camera installed on the bracket of the exterior electronic rearview mirror; environmental modeling based on the ROI area: through obstacle detection in the area of ​​interest where there is a collision risk, an environmental model of both sides of the vehicle body is established, so as to obtain the lateral distance of the obstacle; a time-based collision model: determines the steering control time corresponding to the maximum steering angle and at least one of the collision occurrence time according to the lateral distance of the obstacle and the current motion information, so as to obtain the vehicle warning time and the vehicle braking time; lateral collision warning: performs lateral collision warning control on the vehicle according to the vehicle warning time; lateral collision emergency braking: performs lateral collision braking control on the vehicle according to the vehicle braking time. The present application can further expand the functions of the electronic rearview mirror, and improve the practicality, intelligence level, and competitiveness of the product; the present application can provide early warning and emergency braking for the risk of scratches on both sides, filling the gap in the prior art; the present application uses a monocular depth estimation algorithm, and only needs to use the camera of the external electronic rearview mirror to obtain the three-dimensional environment on both sides of the vehicle body, which can greatly reduce costs; the present application can effectively avoid scratches on both sides of the vehicle body caused by the driver's immature driving skills or lack of concentration, protect the safety of the vehicle body, and thus protect the driver's property safety; the present application can improve the safety of two-wheeled vehicles and pedestrians in the environment on both sides of the vehicle body.

[0099] See also Figure 7 , Figure 7 A block diagram of a vehicle side collision control device according to an embodiment of the present application is shown. The device can be applied to Figure 1 The implementation environment shown is specifically configured in the vehicle 102. The device may also be applicable to other exemplary implementation environments and specifically configured in other devices, and this embodiment does not limit the implementation environment to which the device is applicable.

[0100] like Figure 7 As shown, a vehicle side collision control device 700 according to an embodiment of the present application includes: an acquisition module 701 , a time determination module 702 and a control module 703 .

[0101] The acquisition module 701 is used to acquire the lateral distance of the obstacle and the current motion information of the vehicle, where the lateral distance of the obstacle is used to represent the lateral distance between the obstacles on both sides and the vehicle body when the vehicle starts to turn.

[0102] A time determination module 702, used to determine a collision control time according to current motion information and a lateral distance of an obstacle, wherein the collision control time is used to represent a time corresponding to at least one of avoiding a collision and causing a collision between the vehicle and obstacles on both sides;

[0103] A control module 703, configured to perform at least one of lateral collision warning control and lateral collision braking control on the vehicle based on the collision control time;

[0104] Among them, the lateral distance of the obstacle is obtained by inputting the images of the two sides of the vehicle body to be inspected corresponding to when the vehicle starts to turn into the vehicle body side environment model for obstacle detection. The vehicle body side environment model is trained based on historical vehicle body side images and multiple obstacle label information.

[0105] It should be noted that the vehicle side collision control device provided in the above embodiment and the vehicle side collision control method provided in the above embodiment belong to the same concept, wherein the specific manner in which each module and unit performs the operation has been described in detail in the method embodiment, and will not be repeated here. In actual application, the vehicle side collision control device provided in the above embodiment can distribute the above functions to different functional modules as needed, that is, divide the internal structure of the device into different functional modules to complete all or part of the functions described above, and this is not limited here.

[0106] An embodiment of the present application also provides an electronic device, comprising: one or more processors; a storage device for storing one or more programs, and when the one or more programs are executed by one or more processors, the electronic device implements the vehicle side collision control method provided in the above-mentioned embodiments.

[0107] See also Figure 8 , Figure 8 The structure diagram of the computer system suitable for implementing the electronic device of the embodiment of the present application is shown. It should be noted that: Figure 8 The computer system 800 of the electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0108] like Figure 8 As shown, the computer system 800 includes a central processing unit (CPU) 801, which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM) 802 or the program loaded from the storage part 808 to the random access memory (RAM) 803, such as executing the method in the above embodiment. In the random access memory 803, various programs and data required for system operation are also stored. The central processing unit 801, the read-only memory 802 and the random access memory 803 are connected to each other through a bus 804. The input / output (I / O) interface 805 is also connected to the bus 804.

[0109] The following components are connected to the input / output interface 805: an input section 806 including a keyboard, a mouse, etc.; an output section 807 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker; a storage section 808 including a hard disk, etc.; and a communication section 809 including a network interface card such as a LAN (Local Area Network) card, a modem, etc. The communication section 809 performs communication processing via a network such as the Internet. A drive 810 is also connected to the input / output interface 805 as needed. A removable medium 811, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 810 as needed so that a computer program read therefrom is installed into the storage section 808 as needed.

[0110] In particular, according to an embodiment of the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present application includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a computer program for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication section 809, and / or installed from a removable medium 811. When the computer program is executed by a central processing unit (CPU) 801, various functions defined in the system of the present application are executed.

[0111] It should be noted that the computer-readable medium shown in the embodiment of the present application can be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium can be, for example, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media can include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a flash memory, an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, wherein a computer-readable computer program is carried. This propagated data signal can take a variety of forms, including but not limited to an electromagnetic signal, an optical signal, or any suitable combination of the above. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. A computer program contained on a computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wired, etc., or any suitable combination of the above.

[0112] The flowchart and block diagram in the accompanying drawings illustrate the possible architecture, functions and operations of the system, method and computer program product according to various embodiments of the present application. Wherein, each box in the flowchart or block diagram can represent a module, a program segment, or a part of the code, and the above-mentioned module, program segment, or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flowchart, and the combination of boxes in the block diagram or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0113] The units involved in the embodiments described in the present application can be implemented by software or by hardware, and the units described can also be set in a processor. Among them, the names of these units do not constitute a limitation on the units themselves under certain circumstances. Therefore, the technical solution according to the implementation mode of the present application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, including a number of instructions to enable a computing device (which can be a personal computer, a server, a touch terminal, or a network device, etc.) to execute the method according to the implementation mode of the present application.

[0114] Another aspect of the present application further provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor of a computer, the computer executes the vehicle side collision control method provided in the above-mentioned embodiments. The computer-readable storage medium may be included in the electronic device described in the above-mentioned embodiments, or may exist independently without being assembled into the electronic device.

[0115] In the above embodiments, unless otherwise specified, by using serial numbers such as "first" and "second" to describe common objects, it only means that they refer to different instances of the same object, rather than indicating that the objects being described must adopt a given order, whether in time, space, sorting or any other way.

[0116] The above embodiments are merely illustrative of the principles and effects of the present application, and are not intended to limit the present application. Anyone familiar with the technology may modify or change the above embodiments without violating the spirit and scope of the present application. Therefore, all equivalent modifications or changes made by a person of ordinary skill in the art without departing from the spirit and technical ideas disclosed in the present application shall still be covered by the claims of the present application.

Claims

1. A vehicle side collision control method, characterized in that: The method comprises: Obtaining the lateral distance of the obstacle and the current motion information of the vehicle, wherein the lateral distance of the obstacle is used to represent the lateral distance between the obstacles on both sides and the vehicle body when it is detected that the vehicle starts to turn; Determine a collision control time according to the current motion information and the lateral distance of the obstacle, wherein the collision control time is used to represent a time corresponding to at least one of avoiding a collision and causing a collision between the vehicle and the obstacles on both sides; performing at least one of lateral collision warning control and lateral collision braking control on the vehicle based on the collision control time; The lateral distance of the obstacle is obtained by inputting the images of both sides of the vehicle body to be inspected corresponding to when the vehicle starts to turn into the vehicle body environment model for obstacle detection, and the vehicle body environment model is trained based on historical vehicle body images and multiple obstacle label information; The collision control time includes the steering control time corresponding to the maximum steering angle for the vehicle to avoid collision with the obstacles on both sides, and the collision occurrence time for the vehicle to have redundant collision with the obstacles on both sides; Performing at least one of lateral collision warning control and lateral collision braking control on the vehicle based on the collision control time includes: Determining a vehicle control time according to a preset control time threshold and the collision control time, so as to perform at least one of a lateral collision warning control and a lateral collision braking control on the vehicle, wherein the vehicle control time includes at least one of a vehicle warning time and a vehicle braking time; Determining the vehicle control time according to a preset control time threshold and the collision control time includes: If the collision control time includes the collision occurrence time, the collision occurrence time is used as the collision braking time, and the collision warning time is determined according to a preset collision time threshold and the collision occurrence time; If the steering control time exists in the collision control time, the steering control time is used as the steering braking time, and at least one of the steering warning time is determined according to a preset steering time threshold and the steering control time; If both the collision braking time and the steering braking time exist, the minimum braking time value among the collision braking time and the steering braking time is determined as the vehicle braking time; If both the collision warning time and the steering warning time exist, the minimum warning time value of the collision warning time and the steering warning time is determined as the vehicle warning time.

2. The vehicle side collision control method according to claim 1, characterized in that: Determining the collision control time according to the current motion information and the lateral distance of the obstacle includes: Determining a maximum steering angle for the vehicle to avoid collision with the obstacles on both sides according to the current steering angle, the current vehicle speed and the lateral distance of the obstacle; and / or, Respectively decomposing the current vehicle speed and the current acceleration in the vehicle wheelbase direction to obtain the lateral vehicle speed and the lateral acceleration, and determining the collision occurrence time of the redundant collision between the vehicle and the obstacles on both sides based on the lateral vehicle speed, the lateral acceleration, the difference between the lateral distance of the obstacle and the preset safety distance; The current motion information includes the current steering angle, the current vehicle speed and the current acceleration, and the collision control time includes the collision occurrence time and the steering control time corresponding to the maximum steering angle.

3. The vehicle side collision control method according to claim 2, characterized in that: Performing at least one of lateral collision warning control and lateral collision braking control on the vehicle based on the collision control time includes: Determine a vehicle control time according to a preset control time threshold and the collision control time, wherein the vehicle control time includes at least one of a vehicle warning time and a vehicle braking time; Comparing the collision occurrence time with the vehicle control time to obtain a first comparison result, or comparing the current turning time with the vehicle control time to obtain a second comparison result, wherein the current motion information includes the current turning time; At least one of lateral collision warning control and lateral collision braking control is performed on the vehicle according to the first comparison result, or at least one of lateral collision warning control and lateral collision braking control is performed on the vehicle according to the second comparison result.

4. The vehicle side collision control method according to any one of claims 1 to 3, characterized in that: The method for determining the lateral distance of the obstacle includes: If it is detected that the vehicle starts to turn, images of both sides of the vehicle body to be inspected when the vehicle starts to turn are collected through the exterior electronic rearview mirror; Input the images of both sides of the vehicle body to be inspected into the environment model of both sides of the vehicle body to perform obstacle detection, and obtain the current lateral distance between the obstacles on both sides and the vehicle body; The current lateral distance closest to the vehicle body is used as the lateral distance of the obstacle.

5. The vehicle side collision control method according to any one of claims 1 to 3, characterized in that: Before obtaining the lateral distance of the obstacle of the vehicle, the method further includes: Acquire historical images of both sides of the vehicle and multiple obstacle label information, wherein the historical images of both sides of the vehicle are acquired by an image acquisition device in an exterior electronic rearview mirror; Performing obstacle detection on the region of interest in the historical images on both sides of the vehicle body according to each obstacle label information to obtain actual coordinates of the obstacle, wherein the region of interest is used to characterize the image region where there is a collision risk; Based on the preset device internal parameters of the image acquisition device, the depth of the image corresponding to the region of interest is estimated to obtain a depth image, a three-dimensional grid map corresponding to the depth image is established, and the three-dimensional coordinates of the obstacle in a preset image acquisition coordinate system are obtained according to the three-dimensional grid map and the actual coordinates of the obstacle; The three-dimensional coordinates of the obstacle are converted into a preset vehicle body coordinate system, and the historical lateral distances between the obstacles on both sides and the vehicle body are obtained according to the converted three-dimensional coordinates of the obstacle; The vehicle body side environment model includes obstacle detection, establishment of the three-dimensional grid map and coordinate system conversion training.

6. The vehicle side collision control method according to claim 5, characterized in that: Obstacle detection is performed on the region of interest in the historical images on both sides of the vehicle body according to each obstacle label information to obtain the two-dimensional coordinates of the obstacle, including: Performing image preprocessing on the historical vehicle body side images to obtain first vehicle body side images, wherein the image preprocessing includes performing color restoration on non-vehicle body areas in the historical vehicle body side images by average exposure; Divide the region of interest in the first vehicle body side image, scale the image corresponding to the region of interest according to a preset scaling factor to obtain a second vehicle body side image, and perform pixel filling on the second vehicle body side image to obtain a third vehicle body side image with the same length and width, wherein the region of interest is located in the first region; Performing obstacle detection on the third vehicle body two-side image according to each obstacle tag information to obtain target detection coordinates, and restoring the target detection coordinates to the first vehicle body two-side image according to the preset scaling factor to obtain actual obstacle coordinates; Among them, the first image on both sides of the vehicle body includes the first area located in the middle of the image, the second area and the third area located on both sides of the image, the second area is used to characterize the area where the vehicle body is located, and the third area is used to characterize the area whose distance difference with the second area is greater than or equal to a preset distance threshold.

7. A vehicle side collision control device, characterized in that: The device comprises: An acquisition module, used to acquire the lateral distance of obstacles and current motion information of the vehicle, wherein the lateral distance of obstacles is used to represent the lateral distance between the obstacles on both sides and the vehicle body when it is detected that the vehicle starts to turn; A time determination module, used to determine a collision control time according to the current motion information and the lateral distance of the obstacle, wherein the collision control time is used to represent a time corresponding to at least one of avoiding a collision and causing a collision between the vehicle and the obstacles on both sides; a control module, configured to perform at least one of lateral collision warning control and lateral collision braking control on the vehicle based on the collision control time; The lateral distance of the obstacle is obtained by inputting the images of both sides of the vehicle body to be inspected corresponding to when the vehicle starts to turn into the vehicle body environment model for obstacle detection, and the vehicle body environment model is trained based on historical vehicle body images and multiple obstacle label information; The collision control time includes the steering control time corresponding to the maximum steering angle for the vehicle to avoid collision with the obstacles on both sides, and the collision occurrence time for the vehicle to have redundant collision with the obstacles on both sides; Performing at least one of lateral collision warning control and lateral collision braking control on the vehicle based on the collision control time includes: Determining a vehicle control time according to a preset control time threshold and the collision control time, so as to perform at least one of a lateral collision warning control and a lateral collision braking control on the vehicle, wherein the vehicle control time includes at least one of a vehicle warning time and a vehicle braking time; Determining the vehicle control time according to a preset control time threshold and the collision control time includes: If the collision control time includes the collision occurrence time, the collision occurrence time is used as the collision braking time, and the collision warning time is determined according to a preset collision time threshold and the collision occurrence time; If the steering control time exists in the collision control time, the steering control time is used as the steering braking time, and at least one of the steering warning time is determined according to a preset steering time threshold and the steering control time; If both the collision braking time and the steering braking time exist, the minimum braking time value among the collision braking time and the steering braking time is determined as the vehicle braking time; If both the collision warning time and the steering warning time exist, the minimum warning time value of the collision warning time and the steering warning time is determined as the vehicle warning time.

8. An electronic device, characterized in that: The electronic device comprises: one or more processors; A storage device for storing one or more programs, which, when executed by the one or more processors, enables the electronic device to implement the vehicle side collision control method as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that: A computer program is stored thereon, and when the computer program is executed by a processor of a computer, the computer is caused to execute the vehicle side collision control method according to any one of claims 1 to 6.

Images