Automatic emergency braking control method and device

Abstract

The invention discloses an automatic emergency braking control method and device, and relates to the field of intelligent driving. The method comprises the steps of determining rigid body transformation parameters of a first object between adjacent frames based on an image sequence acquired by a vehicle; acquiring a first image coordinate of a boundary point of a first object in a first image acquired by the vehicle at a first moment; determining a second image coordinate of a boundary point of the first object in the second image based on the rigid body transformation parameter and the first image coordinate; the acquisition moment of the second image is a second moment earlier than the first moment; based on the first image coordinates, the second image coordinates, the first moment and the second moment, determining a mapping relationship between boundary point coordinates of the first object under the own vehicle coordinate system and the acquisition moment; object collision time of the vehicle and the first object at the first moment is determined; based on the object collision time and the mapping relation, determining the position relation between the first object and the vehicle driving path at the collision moment; and controlling an automatic emergency braking function of the vehicle based on the position relation.

Description

Technical Field

[0001] The present disclosure relates to the field of intelligent driving, and in particular to an automatic emergency braking control method and device. Background Art

[0002] In the field of intelligent driving, the AEB (Autonomous Emergency Braking) function plays a vital role in the safety of intelligent driving. This function can control the vehicle to brake automatically in dangerous scenarios where a collision is about to occur to avoid accidents. Most existing solutions filter and estimate the state parameters of the target, such as the speed and position, and then determine whether there is a risk of collision between the target and the vehicle based on the state parameters. For example, it is determined based on the state parameters whether the target is on the vehicle's driving path, that is, whether the target is Inpath. The triggering and suppression of the AEB function can then be determined based on the state parameters of the target estimated by filtering. However, this filtering estimation solution may cause inaccurate state parameters due to sensor detection noise or sensor detection errors, which in turn may lead to incorrect control results for the AEB function. This may cause many problems of false braking in normal scenarios, giving the driver a bad driving experience. Summary of the Invention

[0003] The filtering estimation scheme for Inpath in the related art does not produce accurate results, which may lead to the vehicle's control device failing to control AEB and causing the problem of false braking.

[0004] In order to solve the above technical problems, the present disclosure provides an automatic emergency braking control method and device to make the use of the automatic emergency braking function of a vehicle more accurate and prevent accidental braking.

[0005] An embodiment of the first aspect of the present disclosure provides an automatic emergency braking control method, including: determining a rigid body transformation parameter of a first object between two adjacent frames of images based on an image sequence captured by a vehicle; obtaining a first image coordinate of a boundary point of the first object in a first image captured by the vehicle at a first moment in an image coordinate system; determining a second image coordinate of a boundary point of the first object in a second image in the image sequence in an image coordinate system based on the rigid body transformation parameter and the first image coordinate; the capture moment of the second image is a second moment, and the second moment is earlier than the first moment; determining a mapping relationship between the coordinates of the boundary point of the first object in the vehicle coordinate system and the capture moment of the image in the image sequence based on the first image coordinate, the second image coordinate, the first moment, and the second moment; determining an object collision time between the vehicle and the first object at the first moment; determining a positional relationship between the first object and the vehicle's driving path at the collision moment based on the object collision time and the mapping relationship; and controlling the automatic emergency braking function of the vehicle based on the positional relationship.

[0006] A second embodiment of the present disclosure provides an automatic emergency braking control device, comprising:

[0007] a parameter determination module, configured to determine a rigid body transformation parameter of a first object between two adjacent frames of images based on a sequence of images acquired by the vehicle;

[0008] An acquisition module, configured to acquire a first image coordinate of a boundary point of a first object in a first image captured by the vehicle at a first moment in an image coordinate system;

[0009] a boundary module, configured to determine, based on the rigid body transformation parameters determined by the parameter determination module and the first image coordinates acquired by the acquisition module, second image coordinates of a boundary point of the first object in the second image of the image sequence in the image coordinate system; wherein the acquisition time of the second image is a second moment, and the second moment is earlier than the first moment;

[0010] a mapping module, configured to determine a mapping relationship between the coordinates of a boundary point of the first object in the ego-vehicle coordinate system and the time instant of image sequence acquisition based on the first image coordinates determined by the parameter determination module, the second image coordinates determined by the boundary module, the first moment, and the second moment;

[0011] a collision module, configured to determine a collision time between the vehicle and the first object at a first moment;

[0012] a position determination module, configured to determine a positional relationship between the first object and the vehicle's travel path based on the object collision time determined by the collision module and the mapping relationship determined by the mapping module;

[0013] The processing module is used to control the automatic emergency braking function of the vehicle based on the position relationship determined by the position determination module.

[0014] An embodiment of the third aspect of the present disclosure provides a computer-readable storage medium, which stores a computer program, and the computer program is used to execute the automatic emergency braking control method provided by the embodiment of the first aspect of the present disclosure.

[0015] According to a fourth aspect of the present disclosure, an electronic device is provided, comprising:

[0016] processor;

[0017] a memory for storing instructions executable by the processor;

[0018] The processor is used to read the instructions from the memory and execute the instructions to implement the automatic emergency braking control method provided by the embodiment of the first aspect of the present disclosure.

[0019] An embodiment of the fifth aspect of the present disclosure provides a computer program product, which, when an instruction processor in the computer program product is executed, executes the automatic emergency braking control method provided by the embodiment of the first aspect of the present disclosure.

[0020] In the technical solution provided by the embodiment of the present disclosure, the rigid body transformation parameters of the first object between two adjacent frames of images are obtained by analyzing the images in the image sequence collected by the vehicle. The boundary point coordinates of the first object in all images including the first object in the image sequence can be obtained through the rigid body transformation parameters, and then the mapping relationship between the boundary point coordinates of the first object and the acquisition time can be obtained. Based on the mapping relationship and the object collision time of the first object and the vehicle at the first moment, the positional relationship between the first object and the vehicle's driving path at the moment of collision can be obtained, and the control of the vehicle's automatic emergency braking function can be completed based on the positional relationship. In this technical solution, the rigid body transformation parameters are directly obtained based on the image sequence, which can reflect the correlation relationship between the pixel points corresponding to the first object between adjacent frames and capture more details of the movement of the first object. In this way, the positional relationship between the first object and the vehicle's driving path at the moment of collision can be accurately obtained based on the rigid body transformation parameters, and the automatic emergency braking function of the vehicle can be accurately controlled to reduce the occurrence of false braking. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 It is a schematic diagram of the principle of an automatic emergency braking control method provided by an exemplary embodiment of the present disclosure.

[0022] Figure 2 It is a schematic diagram of an application scenario of an automatic emergency braking control method provided by an exemplary embodiment of the present disclosure.

[0023] Figure 3 It is a schematic diagram of an application scenario of an automatic emergency braking control method provided by another exemplary embodiment of the present disclosure.

[0024] Figure 4 This is a flow chart of an automatic emergency braking control method provided by an exemplary embodiment of the present disclosure. Figure 1 .

[0025] Figure 5 It is a schematic diagram of the principle of the perspective projection theorem provided by an exemplary embodiment of the present disclosure.

[0026] Figure 6 This is a flow chart of an automatic emergency braking control method provided by an exemplary embodiment of the present disclosure. Figure 2 .

[0027] Figure 7 3 is a schematic diagram of collision risk at the moment of collision provided by an exemplary embodiment of the present disclosure.

[0028] Figure 8This is a flow chart of an automatic emergency braking control method provided by an exemplary embodiment of the present disclosure. Figure 3 .

[0029] Figure 9 This is a flow chart of an automatic emergency braking control method provided by an exemplary embodiment of the present disclosure. Figure 4 .

[0030] Figure 10 This is a flow chart of an automatic emergency braking control method provided by an exemplary embodiment of the present disclosure. Figure 5 .

[0031] Figure 11 It is a schematic diagram of the principle of the perspective projection theorem provided by another exemplary embodiment of the present disclosure.

[0032] Figure 12 This is a flow chart of an automatic emergency braking control method provided by an exemplary embodiment of the present disclosure. Figure 6 .

[0033] Figure 13 1 is a structural diagram of an automatic emergency braking control device provided by an exemplary embodiment of the present disclosure.

[0034] Figure 14 2 is a structural diagram of an automatic emergency braking control device provided by another exemplary embodiment of the present disclosure.

[0035] Figure 15 is a structural diagram of an electronic device provided by an exemplary embodiment of the present disclosure. DETAILED DESCRIPTION

[0036] To explain the present disclosure, example embodiments of the present disclosure will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present disclosure, rather than all the embodiments. It should be understood that the present disclosure is not limited to the example embodiments.

[0037] Hereinafter, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of the technical features being referred to. Thus, a feature designated "first" or "second" may explicitly or implicitly include one or more of such features. In the description of this application, unless otherwise specified, "plurality" means two or more. "A and / or B" includes the following three combinations: A only, B only, and a combination of A and B.

[0038] It should be noted that the relative arrangement of components and steps, the numerical expressions and numerical values ​​set forth in these embodiments do not limit the scope of the present disclosure unless specifically stated otherwise.

[0039] Application Overview

[0040] An intelligent driving system is an assisted driving system or autonomous driving system developed based on automated control technology. By sensing and analyzing the vehicle's surroundings, it makes autonomous driving decisions, enabling the vehicle to operate autonomously or semi-autonomously. To ensure that vehicles equipped with intelligent driving systems can safely and reliably complete intelligent driving, vehicles with intelligent driving capabilities are typically equipped with autonomous emergency braking (AEB). This function is used to automatically control the vehicle's braking in dangerous scenarios where a collision is likely, thereby avoiding collisions and ensuring the safety of the vehicle and its passengers.

[0041] To ensure accurate control of the automatic emergency braking function, the related art typically uses filtering to estimate the target's state parameters, such as speed and position. This state parameter is then used to determine whether there is a collision risk between the target and the vehicle. For example, the target is determined to be in the vehicle's path, i.e., whether it is inpath. The filtered estimated target state parameters can then be used to determine whether the automatic emergency braking function should be triggered or inhibited. As can be seen, determining the collision risk between the target and the ego vehicle in the related art relies on direct measurement and filtered estimation of the target's state parameters, such as speed and position. However, the actual driving environment is highly variable, and this variability can interfere with the measurement and estimation of the target's state parameters, resulting in inaccurate measurements and estimates of the target's state parameters. Furthermore, filtered estimation often relies heavily on specific dynamic models (e.g., uniform velocity models, uniform acceleration models, etc.) to predict the target's state parameters. If the model is not correct, the target's state parameter estimates will be inaccurate, leading to an incorrect output regarding whether to inhibit the AEB function.

[0042] In summary, in the related art, determining the collision risk between the vehicle and the target based on the state parameters of the target estimated by filtering cannot accurately control the triggering of the automatic emergency braking function, and there is a safety risk.

[0043] In view of the above problems, the present disclosure provides an automatic emergency braking control method. Figure 1As shown, in this technical solution, by processing the image sequence collected by the vehicle, the rigid body transformation parameters of the first object between adjacent frames are obtained. The rigid body transformation parameters are used to reflect the corresponding relationship between the pixel points corresponding to the first object between adjacent frames. The boundary point coordinates of the first object in the second image collected before the first image are obtained by combining the rigid body transformation parameters and the boundary point coordinates of the first object in the first image collected at the first moment (for example, the current moment) (specifically, the coordinates of the boundary point in the image coordinate system). Based on the acquisition time of all images in the image sequence and the changes in the boundary point coordinates of the first object in each image, the mapping relationship between the boundary point coordinates and the acquisition time in the vehicle coordinate system is obtained. Furthermore, based on the mapping relationship and the predetermined collision time between the vehicle and the first object at the first moment, the positional relationship between the first object and the vehicle's driving path at the collision moment can be obtained. Finally, combined with the positional relationship, accurate control of the vehicle's automatic emergency braking function is completed to prevent false braking.

[0044] In the technical solution provided by the embodiment of the present disclosure, the rigid body transformation parameters of the first object between two adjacent frames of images are obtained by analyzing the images in the image sequence collected by the vehicle. The boundary point coordinates of the first object in all images including the first object in the image sequence can be obtained through the rigid body transformation parameters, and then the mapping relationship between the boundary point coordinates of the first object and the acquisition time can be obtained. Based on the mapping relationship and the object collision time between the first object and the vehicle at the current moment, the positional relationship between the first object and the vehicle's driving path at the moment of collision can be obtained, and the control of the vehicle's automatic emergency braking function can be completed based on the positional relationship. In this technical solution, the rigid body transformation parameters are directly obtained based on the image sequence, which can reflect the correlation relationship between the pixel points corresponding to the first object between adjacent frames and capture more details of the movement of the first object. In this way, the positional relationship between the first object and the vehicle's driving path at the moment of collision can be accurately obtained based on the rigid body transformation parameters, and the automatic emergency braking function of the vehicle can be accurately controlled to reduce the occurrence of false braking.

[0045] Exemplary Systems

[0046] Figure 2 This is an exemplary application scenario of the automatic emergency braking control method provided by the present disclosure.

[0047] like Figure 2 As shown, the scene includes a driving intelligent driving vehicle 201 and a first object 202 in front of the intelligent driving vehicle. First object 202 can be one or more, and can be another driving vehicle, a walking pedestrian, a two-wheeled or three-wheeled vehicle, or other obstacles in the vehicle's path. This disclosure does not impose specific limitations on this.

[0048] The intelligent driving vehicle 201 is equipped with an image acquisition device for capturing images of the environment while the intelligent driving vehicle 201 is traveling. The image of the environment may include a first object 202 surrounding the vehicle. For example, the image acquisition device may be a camera for capturing images directly in front of the intelligent driving vehicle 201, or a camera for capturing images in front of the left side of the intelligent driving vehicle 201. This disclosure does not impose any specific limitations on this.

[0049] The intelligent driving vehicle 201 may also be provided with a control device for executing the automatic emergency braking control method provided by the embodiments of the present disclosure. The control device may specifically be an onboard computer or an intelligent driving system in the intelligent driving vehicle 201. This disclosure does not impose any specific restrictions on this.

[0050] Specifically, the control device of the intelligent driving vehicle 201 can process the collected image sequence to obtain rigid body transformation parameters that can reflect the correlation relationship between the pixel points corresponding to the first object between adjacent frames, and based on the rigid body transformation parameters and the boundary point coordinates of the first object in the first image collected at the first moment (for example, the current moment), obtain the boundary point coordinates of the first object in other images including the first object in the image sequence other than the first image. The control device of the intelligent driving vehicle 201 can obtain the mapping relationship between the image collection time and the boundary point coordinates of the first object based on the collection time and corresponding boundary point coordinates of all images including the first object in the image sequence, and then predict the positional relationship between the first object and the vehicle's driving path at the time of collision. Finally, the control device of the intelligent driving vehicle 201 can complete the control of the automatic emergency braking function based on the positional relationship, so that the use of the vehicle's automatic emergency braking function is more accurate and prevents false braking.

[0051] Figure 3 This is an exemplary application scenario of the automatic emergency braking control method provided by the present disclosure.

[0052] like Figure 3 As shown, the scene includes a driving intelligent driving vehicle 301, a server device 303, and a first object 302 in front of the intelligent driving vehicle 301. The specific definition of the first object 302 can be referred to the relevant description of the first object 202 in the aforementioned embodiment and is not repeated here. The server device 303 stores the image sequence captured by the intelligent driving vehicle 301, as well as any other possible data. Data transmission between the server device 303 and the intelligent driving vehicle 301 can be achieved through any possible wired or wireless communication means.

[0053] Among them, the intelligent driving vehicle 301 is provided with an image acquisition device. The functions of the image acquisition device can refer to the image acquisition device provided on the intelligent driving vehicle 201 in the aforementioned embodiment, and will not be repeated here. The intelligent driving vehicle 301 can also be provided with a control device. The functions of the control device can refer to the control device provided on the intelligent driving vehicle 201 in the aforementioned embodiment, and will not be repeated here. The difference between the two is that the control device in the embodiment of the present disclosure needs to obtain the image sequence from the server device first when it needs to process the image sequence.

[0054] It should be noted that when the application scenario of the automatic emergency braking control method provided in the present disclosure includes an intelligent driving vehicle and a server-side device, in addition to being implemented solely by the control device on the intelligent driving vehicle, the method can also be implemented partially or completely on the server-side device.

[0055] Exemplary Methods

[0056] The automatic emergency braking control method provided by the embodiments of the present disclosure is introduced below with reference to the accompanying drawings.

[0057] Figure 4 This is a flow chart of an automatic emergency braking control method provided by an exemplary embodiment of the present disclosure. This embodiment can be applied to intelligent driving vehicles with automatic emergency braking functions, such as Figure 4 As shown, the method may include S401-S404:

[0058] S401 : Determine rigid body transformation parameters of a first object between two adjacent frames of images based on an image sequence collected by a vehicle.

[0059] In an embodiment of the present disclosure, when the vehicle is driving, the image acquisition device / image sensor (such as a front camera) on the vehicle can capture images of the area in front of the vehicle in real time to obtain an image sequence consisting of multiple frames of images. The image sequence can be stored in the local memory of the vehicle or stored on a server device that is in communication with the vehicle. The present disclosure does not impose any specific restrictions on this. If there is a first object in front of the vehicle, an image including the first object may be present in the image sequence. When the technical solution provided by an embodiment of the present disclosure is implemented, in order to save computing power, only the image frames including the first object in the image sequence can be calculated and processed. In addition, if computing power is not considered, in another embodiment of the present disclosure, all images in the image sequence can also be processed. However, during the processing, if the first object is not detected in the image to be processed, the image can be skipped and the next frame of the image can be processed until the next frame of the image including the first object is processed to obtain the corresponding rigid body transformation parameters.

[0060] Furthermore, in some embodiments, the image sequence mentioned in the present disclosure may also consist of only images including the first object, specifically, images including the first object among all images captured by the image capture device may be sorted according to the capture time.

[0061] In the disclosed embodiment, the rigid body transformation parameters are used to characterize the correlation between the pixels corresponding to the first object between two adjacent image frames. In one possible implementation, the correlation between the pixels corresponding to the first object between two adjacent image frames, i.e., the rigid body transformation parameters, can be obtained using optical flow tracking.

[0062] In an embodiment of the present disclosure, when there are images in the image sequence that do not include the first object, and the images that include the first object are not all continuous, the two adjacent frames of images (or adjacent frames) in the embodiment of the present disclosure refer to the two adjacent frames of images in the sub-image sequence that includes the first object in the image sequence. For example, if the image sequence includes the following five consecutively captured images: image 1, image 2, image 3, image 4, and image 5, where image 1, image 3, and image 4 include the first object, then image 1 and image 3 are two adjacent frames of images in the embodiment of the present disclosure, and image 3 and image 4 are also two adjacent frames of images in the embodiment of the present disclosure. The same applies to subsequent embodiments.

[0063] In addition, when there are only images including the first object in the image sequence, the two adjacent frames of images in the embodiment of the present disclosure refer to two adjacent frames of images in the image sequence.

[0064] Among them, optical flow tracking is a computer vision technology that estimates the position and motion trajectory of an object by analyzing the pixel motion between consecutive image frames. Its implementation is mainly based on the following three assumptions: (1) Brightness constancy assumption: the surface brightness of the same object remains unchanged between consecutive frames (that is, the grayscale value of the pixel does not change with motion). (2) Spatial consistency assumption: the motion trends of adjacent pixels are similar. (3) Temporal continuity assumption: the time interval between frames is extremely short and the object displacement is small. Specifically, the process of optical flow tracking to obtain the rigid body transformation parameters of the first object in the image between adjacent frames is as follows: by processing the adjacent frames in the image sequence with the optical flow algorithm, the displacement vector (or optical flow field) of the pixel points of the first object in the adjacent frames is obtained. The key point coordinates of the first object in the next frame of the adjacent frame are obtained by the displacement vector and the key point coordinates of the first object in the previous frame of the adjacent frame (specifically, the coordinates of the key point in the image coordinate system). Based on the key point coordinates of the first object in the adjacent frames, the rigid body transformation parameters of the first object between adjacent frames are determined. Among them, the process of obtaining the displacement vector of the pixel point can also be called optical flow estimation, and optical flow tracking is a further processing completed on the basis of optical flow estimation.

[0065] Based on this, in some embodiments, in order to obtain the rigid body transformation parameters of the first object between two adjacent frames of images, the vehicle control device can perform optical flow tracking on the first object in the image sequence, thereby determining the rigid body transformation parameters of the first object between the two adjacent frames of images. Specifically, the vehicle control device can use the aforementioned optical flow tracking method to perform optical flow tracking on the first object in the previous frame of the two adjacent frames of images each time the image acquisition device captures two adjacent frames of images, thereby obtaining the rigid body transformation parameters of the first object between the adjacent frames. Since the rigid body transformation parameters are directly based on the image data and obtained by optical flow tracking, they can capture more details of the object's movement with high accuracy, making the subsequent mapping relationship gradually obtained based on the rigid body transformation parameters and the positional relationship between the first object and the vehicle's driving path at the time of collision more accurate, and smaller than the error of the positional relationship obtained by filtering estimation. Combined with this positional relationship, the control of the automatic emergency braking function can be completed more accurately, reducing the occurrence of false braking.

[0066] Typically, when optical flow tracking of a first object in an image is required, the first object must first be detected for the first frame in the image sequence that includes the first object, resulting in a detection frame that encompasses the first object. Because the detection frame has a fixed shape and a clear position, the optical flow tracking algorithm can more quickly track the first object within the detection frame, thereby continuously obtaining detection frames for the first object in subsequent frames. Furthermore, to prevent tracking drift (a discrepancy between the tracked detection frame and the true position of the target object) or tracking failure (loss of the target object and inability to track a new detection frame) during optical flow tracking, in an embodiment of the present disclosure, an object detection algorithm is applied to the previous frame in all adjacent frames to detect the first object and determine a more accurate detection frame. In an embodiment of the present disclosure, the detection frame can specifically be a rectangular frame. The object detection algorithm can be any possible object detection model, such as a YOLO (you only look once) model, an SSD (single shot multibox detector) model, a DETR (detection transformer) model, etc., and this disclosure does not impose any specific limitations on this.

[0067] On this basis, in a possible implementation method, performing optical flow tracking on the first object in the image sequence and determining the rigid body transformation parameters of the first object between two adjacent frames of images can specifically include: performing optical flow tracking on the first feature point in the first detection frame of the first object in the previous frame of the two adjacent frames of images in the image sequence, and obtaining the coordinates of the second feature point corresponding to the first feature point in the next frame of the two adjacent frames of images; establishing a fitting equation to be solved based on the coordinates of the first feature point and the coordinates of the second feature point; and solving the rigid body transformation parameters in the fitting equation to be solved.

[0068] The coordinates of the feature points (first feature points or second feature points) refer to the coordinates of the feature points in the image coordinate system. In some possible implementations, the fitting equations to be solved may include N of the following equation groups:

[0069] x′=s×x+dx;(1)

[0070] y′=s×y+dy;(2)

[0071] Among them, N is the number of first feature points, x′ is the horizontal coordinate of the second feature point, y′ is the vertical coordinate of the second feature point, x is the horizontal coordinate of the first feature point, y is the vertical coordinate of the first feature point, s is the scaling factor in the rigid body transformation parameters, dx is the horizontal translation in the rigid body transformation parameters, and dy is the vertical translation in the rigid body transformation parameters.

[0072] After establishing the fitting equation to be solved, any feasible cost function can be set up and the linear regression algorithm can be used to solve the fitting parameters to be solved. For example, the cost function can be:

[0073]

[0074] Among them, x cur i is the horizontal coordinate of the second feature point i obtained by optical flow tracking, x′ i is the horizontal coordinate of the second feature point i predicted based on the rigid body transformation parameters, y cur i is the true vertical coordinate of the second feature point i obtained by optical flow tracking, y′ i is the ordinate of the i-th second feature point predicted based on the rigid body transformation parameters.

[0075] In this way, the rigid body transformation parameters of the first object in the image sequence between two adjacent frames can be obtained. In addition, the significance of obtaining the rigid body transformation parameters for multiple feature points is that: since the displacement vectors of the associated feature points obtained by optical flow tracking will be affected by the brightness of the associated feature points themselves, the accuracy of the displacement vectors of different associated feature points is different. Therefore, the displacement vectors of any pair of associated feature points cannot represent the overall displacement change of the pixel points between the first image and the second image. Therefore, when the displacement vectors and corresponding coordinates of multiple pairs of associated feature points are obtained, in order to more accurately determine the second image coordinates of the boundary points of the first object in the second image in the image coordinate system, it is necessary to determine the rigid body transformation parameters that can accurately represent the overall displacement change of the pixels between the first image and the second image based on the coordinates of the multiple pairs of associated feature points. Further, based on the rigid body transformation parameters, a more accurate second image coordinate of the boundary points of the first object in the second image in the image coordinate system can be obtained.

[0076] S402 : Obtain first image coordinates of a boundary point of a first object in a first image captured by a vehicle at a first moment in an image coordinate system.

[0077] In some embodiments, optical flow tracking can be used to accurately obtain the first image coordinates. Therefore, S402 may specifically include performing optical flow tracking on the first object in a third image in the image sequence that was captured earlier than the first image to obtain the first image coordinates. Furthermore, as described above after S401, when performing optical flow tracking on the first object, optical flow tracking is specifically performed on the detection frame of the first object. Furthermore, when determining the first image coordinates of the boundary points of the first object in the first image through optical flow tracking, it is necessary to ensure that the capture times of the third image and the first image are as close as possible. This minimizes the change in ambient light between the first and third images, and reduces the displacement of the first object. Therefore, the third image can be an image that includes the first object in the frame preceding the first image, meaning that the first and third images are adjacent frames. This applies similarly to subsequent embodiments. Of course, if the vehicle's image capture device captures images at a high frame rate and the interval between adjacent frames is very short, the first and third images may not be adjacent frames, meaning that they may be spanned images separated by multiple frames.

[0078] In addition, in reality, the shape of the first object itself is irregular, and its boundary points are difficult to define. The shape of the detection frame is fixed, so the boundary points of the detection frame can be used as the boundary points of the first object. Taking the detection frame as a rectangular frame as an example, the boundary points of the detection frame can specifically include the upper left corner point and the upper right corner point, and / or, the lower left corner point and the lower right corner point. Of course, the boundary points of the first object can also be feature points on the first object or derived based on the feature points on the first object. Specifically, based on the distribution of feature points on the first object, as well as the shape and size of the first object, the pixel points in the image that can serve as the boundary points of the first object or the feature points that can serve as the boundary points of the first object can be determined.

[0079] In this way, optical flow tracking can accurately obtain the first image coordinates of the boundary points of the first object in the first image captured at the first moment based on image information. Furthermore, based on the first image coordinates, the mapping relationship between the coordinates of the boundary points of the first object and the image capture time can be accurately determined, as well as the positional relationship between the first object and the vehicle's travel path at the moment of collision. This allows the vehicle's control device to accurately control the automatic emergency braking function based on this positional relationship, reducing the occurrence of false braking.

[0080] S403 : Determine second image coordinates of a boundary point of the first object in a second image in the image sequence in the image coordinate system based on the rigid body transformation parameters and the first image coordinates.

[0081] The acquisition time of the second image is the second moment, and the second moment is earlier than the first moment.

[0082] Exemplarily, after obtaining the rigid body transformation parameters and first image coordinates of the first object between two adjacent frames in the image sequence, the second image coordinates of the boundary point of the first object in the second image can be obtained based on the following formula.

[0083]

[0084] Among them, x i is the horizontal coordinate of the first image coordinate of the latter frame in two adjacent frames of images, y i is the vertical coordinate of the first image coordinate of the latter frame in two adjacent frames of images, s i-1 is the scaling factor in the rigid body transformation parameters of the first object between the i-th frame and the i-1-th frame, dx i-1 is the horizontal translation of the first object in the rigid body transformation parameters between the i-th frame and the i-1-th frame, dy i-1 is the vertical translation of the first object in the rigid body transformation parameters between the i-th frame and the i-1-th frame. The i-th frame is the next frame in the two adjacent frames, and the i-1-th frame is the previous frame in the two adjacent frames. i-1is the horizontal coordinate of the first image coordinate of the i-1th frame in the image sequence, y i-1 is the vertical coordinate of the first image coordinate of the i-1th frame in the image sequence.

[0085] S404 : Determine a mapping relationship between the coordinates of the boundary point of the first object in the vehicle coordinate system and the capture time of the images in the image sequence based on the first image coordinates, the second image coordinates, the first time, and the second time.

[0086] The technical solution provided by the embodiments of this disclosure aims to determine whether a first object is in the vehicle's travel path at the moment of collision. To this end, it is necessary to first determine the coordinates of the first object's boundary points at the moment of collision. To obtain the coordinates of the first object's boundary points at the moment of collision, it is necessary to determine the mapping relationship between the coordinates of the first object's boundary points in the ego-vehicle coordinate system and the time of image acquisition.

[0087] Therefore, after executing S403, the image coordinates of the boundary points of the first object in all images including the first object in the image sequence and the acquisition time of all images in the image sequence are obtained, and the mapping relationship between the boundary points of the first object and the acquisition time of the image can be obtained based on these data.

[0088] In some embodiments, to obtain a mapping relationship, the image coordinates (first image coordinates and second image coordinates) must first be converted to coordinates in the ego-vehicle coordinate system before determining the mapping relationship. Specifically, based on the relationship between the image coordinate system and the camera coordinate system, the image coordinates can be converted to camera coordinates in the camera coordinate system. Furthermore, based on the relationship between the camera coordinate system and the ego-vehicle coordinate system, the camera coordinates can be converted to ego-vehicle coordinates in the ego-vehicle coordinate system.

[0089] S405: Determine the collision time between the vehicle and the first object at the first moment.

[0090] In the embodiment of the present disclosure, S405 may be executed at any possible time before S406. Figure 4 The execution order shown in is only an example.

[0091] In the embodiment of the present disclosure, the object collision time between the vehicle and the first object at the first moment is the time predicted at the first moment when the object and the vehicle may collide, which indicates that if the vehicle and the first object travel according to the parameters at the predicted collision time, starting from the first moment, the vehicle and the first object will collide after the collision time evolves.

[0092] In the disclosed embodiments, the object collision time can be derived using either filtering estimation or optical flow tracking. The object collision time derived using filtering estimation is referred to as the numerical domain collision time, while the object collision time derived using optical flow tracking is referred to as the image domain collision time.

[0093] In some embodiments, the object collision time is obtained by optical flow tracking as follows: optical flow tracking is performed on the first object in the third image in the image sequence whose acquisition time is earlier than the first image, and the image domain collision time between the vehicle and the first object at the first moment is obtained, and the image domain collision time between the vehicle and the first object at the first moment is determined as the object collision time.

[0094] In conjunction with the relevant statements after S401 in the above embodiment, it can be seen that when performing optical flow tracking on the first object, optical flow tracking is specifically performed on the detection frame of the first object. Based on this, using optical flow tracking to obtain the image domain collision time may include the following steps:

[0095] X1. Perform optical flow tracking on the first detection frame of the first object in the third image to obtain a height of the second detection frame of the first object in the first image.

[0096] Exemplarily, X1 may specifically include: performing optical flow tracking on the first feature point in the first detection frame of the first object in the third image to obtain the rigid body transformation parameters of the first object between the third image and the first image; determining the vertex coordinates of the second detection frame of the first object in the first image based on the rigid body transformation parameters and the vertex coordinates of the first detection frame; and determining the height of the second detection frame based on the vertex coordinates of the second detection frame.

[0097] Determining the vertex coordinates of the first detection frame may include: detecting the first object in the third image using an object detection algorithm to obtain a first detection frame of the first object, and obtaining the vertex coordinates of the first detection frame based on the first detection frame. The specific implementation of determining the rigid body transformation parameters can refer to the relevant descriptions in the aforementioned embodiments and will not be repeated here. After obtaining the rigid body transformation parameters, the vertex coordinates of the second detection frame can be obtained by combining formulas (1) and (2) in the aforementioned embodiments.

[0098] X2. Determine the image domain collision time between the vehicle and the first object at the first moment based on the height of the first detection frame, the height of the second detection frame, and the time difference between the first moment and the capture moment of the third image.

[0099] In the disclosed embodiment, the first and third images are adjacent frames or inter-frame images in an image sequence captured in real time by the vehicle's image acquisition device. Therefore, the time difference between the two corresponding capture times is very short, and the motion between the two during this time difference is minimal. Therefore, it can be assumed that during this time difference, the vehicle and the first object are moving at a constant velocity or with a constant acceleration relative to each other in the longitudinal direction. The longitudinal direction is the direction of travel specified by the vehicle's path.

[0100] In some embodiments, assuming that the vehicle and the first object move at a constant speed relative to each other in the longitudinal direction within the acquisition time difference, if the longitudinal distance between the vehicle and the first object at the first moment is D1, then:

[0101] D1=V×T1;(6)

[0102] Where V is the relative velocity of the uniform relative motion between the vehicle and the first object, and T1 is the time from the first moment when the vehicle and the first object meet in the longitudinal direction, which can be called the current collision time. During vehicle travel, the motion of the vehicle and the first object has longitudinal and lateral components, that is, the velocity has longitudinal and lateral components, and the acceleration also has longitudinal and lateral components. Referring to the vehicle coordinate system (VCS), the longitudinal direction is the direction of vehicle travel, and the lateral direction is perpendicular to the longitudinal direction. The aforementioned relative velocity refers to the difference between the longitudinal component of the vehicle's velocity and the longitudinal component of the first object's velocity.

[0103] If the longitudinal distance between the vehicle and the first object at the time of collecting the third image is D0, then:

[0104] D0=V×T0; (7)

[0105] T0 is the time when the vehicle and the first object meet in the longitudinal direction from the moment the third image is captured, which can be called the historical collision time.

[0106] Further, refer to Figure 5 As shown, based on the perspective projection theorem, we can know the following formula:

[0107]

[0108] Wherein, h is the pixel height of the first object in the image, f is the focal length of the image acquisition device on the vehicle, H is the actual height of the first object, and D is the longitudinal distance between the image acquisition device and the first object. Figure 5 The viewpoint in the principle diagram of the perspective projection theorem shown is specifically the optical center of the camera, and the imaging plane is moved from the side of the viewpoint away from the first object to between the viewpoint and the first object for the convenience of calculation.

[0109] Substituting formula (8) into formula (6) and formula (7), we can obtain:

[0110]

[0111] Here, h0 is the height of the first detection frame of the first object in the third image, and h1 is the height of the second detection frame of the first object in the first image.

[0112] In addition, the following relationships exist:

[0113] T1=T0+Δt;(11)

[0114] Wherein, Δt is the acquisition time difference between the acquisition time of the first moment and the acquisition time of the third image.

[0115] The simultaneous combination of formulas (9)-(11) can be used to obtain the formula for calculating the first collision time between the vehicle and the first object at the time of acquisition of the third image:

[0116]

[0117] Thus, the first collision time can be calculated using the above formula (12). Subsequently, the first collision time can be determined as the image domain collision time.

[0118] Of course, the above example assumes that the vehicle and the first object move at a constant speed relative to each other in the longitudinal direction within the acquisition time difference. If it is assumed that the vehicle and the first object move at a constant acceleration relative to each other in the longitudinal direction within the acquisition time difference, the calculation formulas for longitudinal distance (i.e., formulas (6) and (7)) in the above calculation formulas should be adjusted to displacement calculation formulas for uniformly accelerated motion, and the remaining formulas should be adaptively adjusted.

[0119] In addition, the first collision time obtained using the above formula (12) is a single-frame collision time obtained after optical flow tracking of the third image of a single frame. However, the single-frame collision time may produce errors due to the influence of noise in the optical flow tracking process. In order to reduce the error of the first collision time, the first collision time can be smoothed based on the image domain collision time corresponding to the historical frame, thereby determining a more accurate image domain collision time between the vehicle and the first object at the first moment. The image domain collision time corresponding to the historical frame may include the image domain collision time between the vehicle and the first object at the third moment. The third moment is before the first moment, and the third moment may include multiple acquisition moments before the first moment.

[0120] Of course, the image domain collision time corresponding to each historical frame is also a more accurate image domain collision time after smoothing.

[0121] In the embodiments of the present disclosure, the specific implementation of smoothing the first collision time using the image domain collision time corresponding to the historical frame can be implemented in any feasible manner, such as adjusting the first collision time based on the mapping relationship between the image domain collision time corresponding to all images (including multiple frames of historical images (i.e., images captured at the third moment) and the first image) and the capture moment. The present disclosure does not impose specific restrictions on this.

[0122] In this way, optical flow tracking can accurately determine the image-domain collision time based on the image information and use it as the object collision time. Because the image-domain collision time is directly derived based on the pixel height change of the first object in the image data, it can capture more details of the object's motion with minimal error. Therefore, the subsequent positional relationship between the first object and the vehicle's travel path at the moment of collision determined based on the image-domain collision time is also more accurate, which in turn allows for better control of the automatic emergency braking function and reduces the occurrence of false braking.

[0123] Furthermore, in some embodiments, optical flow tracking may fail to complete or determine an accurate image-domain collision time at certain times due to sudden changes in ambient light or occlusion of the first object by other objects in the environment. In such cases, the image-domain collision time may be deemed abnormal, and the numerical-domain collision time between the vehicle and the first object at the first moment, obtained using a filter estimation method, may be used to determine the object collision time. This allows the object collision time between the vehicle and the first object at the first moment to be determined even when the image-domain collision time is unavailable or inaccurate.

[0124] S406 : Determine the positional relationship between the first object and the vehicle's travel path at the moment of collision based on the object collision time and the mapping relationship.

[0125] In some embodiments, after obtaining the mapping relationship, the coordinates of the object boundary point of the first object in the vehicle coordinate system at the moment of collision can be obtained based on the mapping relationship and the object collision time, and then based on the coordinates of the vehicle's boundary point in the vehicle coordinate system and the object boundary point coordinates, it can be determined whether the first object is in the vehicle's driving path at the moment of collision.

[0126] Based on this, in some embodiments, combined with Figure 4 , refer to Figure 6 As shown, S406 may specifically include S4061 and S4062:

[0127] S4061: Determine the coordinates of the object boundary point of the first object in the vehicle coordinate system at the moment of collision based on the mapping relationship and the object collision time.

[0128] The purpose of the technical solution provided by the embodiment of the present disclosure is to determine whether the first object is in the driving path of the vehicle at the time of collision, that is, the positional relationship between the first object and the vehicle at the time of collision. And this positional relationship is mainly determined based on the lateral coordinates (specifically the Y-axis coordinates) of the boundary point of the first object in the self-vehicle coordinate system. Therefore, in the embodiment of the present disclosure, the mapping relationship may include a sub-mapping relationship between the lateral coordinates (specifically the Y-axis coordinates) of the boundary point of the first object in the self-vehicle coordinate system and the acquisition time. Since the lateral movement of the first object is slow and approximately uniform, the coordinates of the boundary point of the first object in the self-vehicle coordinate system are most likely linearly related to time. Therefore, the sub-mapping relationship can be represented by a linear function.

[0129] In the embodiment of the present disclosure, taking the left boundary point (lower left vertex / upper left vertex) and the right boundary point (lower right vertex / upper right vertex) of the detection box of the object as an example, the sub-mapping relationship corresponding to the left boundary point can be expressed as l =k l t+b l Representation, the sub-mapping relationship corresponding to the right boundary point can be represented by X r =k r t+b characterization. Among them, X l is the horizontal coordinate of the left boundary point in the vehicle coordinate system at time t, X r is the horizontal coordinate of the right boundary point in the vehicle coordinate system at time t. t = 0 is the first moment.

[0130] Substituting the collision time (t) into the two linear functions representing the sub-mapping relationship yields the coordinates of the first object's boundary points in the ego-vehicle coordinate system at the time of collision. These coordinates specifically include the Y-axis coordinates of the first object's boundary points (left and right) in the ego-vehicle coordinate system.

[0131] For example, refer to Figure 7 As shown, with t as the horizontal coordinate and X as the vertical coordinate, the two sub-mapping relationships can be represented by two parallel or approximately parallel straight lines, which can be called the left boundary line and the right boundary line.

[0132] S4062: Determine a positional relationship between the first object and the vehicle's travel path at the moment of collision based on the coordinates of the vehicle's boundary point in the vehicle coordinate system and the coordinates of the first object's boundary point in the vehicle coordinate system at the moment of collision.

[0133] When the coordinates of the boundary point of the first object in the ego-vehicle coordinate system at the moment of collision are obtained, it can be determined whether the coordinates of the boundary point of the object are within the boundary range of the vehicle. The boundary range of the vehicle is determined based on the coordinates of the boundary point of the vehicle in the ego-vehicle coordinate system.

[0134] Based on this, in some embodiments, combined with Figure 6 , refer to Figure 8 As shown, S4062 may specifically include S801-S803:

[0135] S801: Determine a boundary range of the vehicle in the vehicle coordinate system at the moment of collision based on the coordinates of the boundary points of the vehicle in the vehicle coordinate system.

[0136] The boundary points of the vehicle may refer to two points on the vehicle body that are farthest apart in the transverse direction, such as the outermost points on two rearview mirrors.

[0137] For example, after determining the coordinates of the boundary points of the vehicle in the vehicle coordinate system, the control device of the vehicle refers to Figure 7 As shown in , the trajectory line of the vehicle boundary range changing with time can be determined based on the Y-axis coordinate of the vehicle boundary point in the vehicle coordinate system, and then the boundary range of the vehicle in the vehicle coordinate system at the time of collision can be obtained. Figure 7 The vehicle boundary line shown corresponds to the X range at the time of collision.

[0138] After determining the boundary range of the vehicle in the ego-vehicle coordinate system at the moment of collision, the positional relationship between the first object and the vehicle's travel path at the moment of collision can be determined based on the coordinates of the object boundary point of the first object in the ego-vehicle coordinate system.

[0139] Specifically, if the coordinates of the left boundary point or the right boundary point in the object boundary point coordinates are within the boundary range of the vehicle in the ego vehicle coordinate system at the time of collision, S802 is executed. For example, if the coordinates of the left boundary point and the right boundary point in the object boundary point coordinates are (ttc, -0.1) and (ttc, -0.3), respectively, and the boundary range of the vehicle in the ego vehicle coordinate system at the time of collision is (ttc, -0.12) to (ttc, 0.12), then the coordinates of the left boundary point in the object boundary point coordinates (ttc, -0.1) are within the boundary range, and S802 is executed.

[0140] If the coordinates of the left and right boundary points in the object boundary point coordinates are not within the boundary range of the vehicle in the ego vehicle coordinate system at the time of collision, then S803 is executed. For example, if the coordinates of the left and right boundary points in the object boundary point coordinates are (ttc, -0.15) and (ttc, -0.3), respectively, and the boundary range of the vehicle in the ego vehicle coordinate system at the time of collision is (ttc, -0.12) to (ttc, 0.12), then the coordinates of the left and right boundary points in the object boundary point coordinates are not within the boundary range, and S803 is executed.

[0141] S802: In response to the coordinates of the boundary point of the object being within the boundary range, determining that the positional relationship is that the first object is in the driving path of the vehicle at the moment of collision.

[0142] S803: In response to the coordinates of the boundary point of the object being outside the boundary range, determining that the positional relationship is that the first object is not in the driving path of the vehicle at the time of collision.

[0143] It should be noted that, in the embodiment of the present disclosure, in one execution process of the automatic emergency braking control method, S802 and S803 are executed selectively.

[0144] In this way, through the coordinates of the vehicle's boundary points in the vehicle's coordinate system and the coordinates of the object boundary points of the first object in the vehicle's coordinate system at the moment of collision, it is possible to accurately determine whether the first object will be in the vehicle's driving path at the moment of collision, that is, the positional relationship between the first object and the vehicle at the moment of collision, and then accurately control the automatic emergency braking function based on the positional relationship.

[0145] Based on the technical solution provided by the embodiment of the present disclosure, through the mapping relationship between the boundary point coordinates of the first object and the acquisition time, as well as the object collision time between the vehicle and the first object at the first moment, it can be determined whether the first object will be on the vehicle's driving path at the moment of collision, that is, the positional relationship between the first object and the vehicle at the moment of collision. Since the mapping relationship is derived based on the image information in the image sequence, and the image information can intuitively and accurately reflect the details of the object's movement, the mapping relationship can accurately reflect the changes in the boundary point coordinates of the first object with the acquisition time. Furthermore, the positional relationship determined based on the mapping relationship is also more accurate, and the automatic emergency control function can be accurately controlled based on the positional relationship to reduce the occurrence of false braking.

[0146] S407: Control the automatic emergency braking function of the vehicle based on the position relationship.

[0147] In the technical solution provided by the embodiment of the present disclosure, the rigid body transformation parameters of the first object between two adjacent frames of images are obtained by analyzing the images in the image sequence collected by the vehicle. The boundary point coordinates of the first object in all images including the first object in the image sequence can be obtained through the rigid body transformation parameters, and then the mapping relationship between the boundary point coordinates of the first object and the acquisition time can be obtained. Based on the mapping relationship and the object collision time of the first object and the vehicle at the first moment, the positional relationship between the first object and the vehicle's driving path at the moment of collision can be obtained, and then the control of the vehicle's automatic emergency braking function can be completed based on the positional relationship. In this technical solution, the rigid body transformation parameters are directly obtained based on the image sequence, which can reflect the correlation relationship between the pixel points corresponding to the first object between adjacent frames and capture more details of the movement of the first object. In this way, the positional relationship between the first object and the vehicle's driving path at the moment of collision can be accurately obtained based on the rigid body transformation parameters, and the automatic emergency braking function of the vehicle can be accurately controlled to reduce the occurrence of false braking.

[0148] In some embodiments, when determining the mapping relationship in S404, the image coordinates (first image coordinates and second image coordinates) need to be converted into coordinates in the vehicle coordinate system before determining the mapping relationship. Figure 4 , refer to Figure 9 As shown, S404 may specifically include S4041 and S4042:

[0149] S4041 : Convert the first image coordinates into first ego-vehicle coordinates in the ego-vehicle coordinate system, and convert the second image coordinates into second ego-vehicle coordinates in the ego-vehicle coordinate system.

[0150] In the embodiment of the present disclosure, the first image coordinates may be converted into the first ego-vehicle coordinates, and the second image coordinates may be converted into the second ego-vehicle coordinates based on the conversion relationship between the image coordinate system and the ego-vehicle coordinate system.

[0151] S4042: Determine a mapping relationship between the coordinates of the boundary point of the first object in the ego-vehicle coordinate system and the capture time of the image in the image sequence based on the first ego-vehicle coordinate system, the second ego-vehicle coordinate system, the first time, and the second time.

[0152] In one possible implementation, after obtaining the first ego-vehicle coordinates, the second ego-vehicle coordinates, the first time instant, and the second time instant, a least-squares fitting algorithm can be used to obtain a first function representing the relationship between the coordinates of the left boundary point of the first object and the acquisition time instant, and a second function representing the relationship between the coordinates of the right boundary point of the first object and the acquisition time instant. These two functions may be two sub-mappings included in the mapping relationship between the coordinates of the boundary points of the first object and the acquisition time instant. Examples of the first and second functions can be found in the description after S4061 in the aforementioned embodiment and are not further described here.

[0153] The technical solution provided by the disclosed embodiments uses a conversion relationship between the image coordinate system and the vehicle coordinate system to convert the first and second image coordinates into first and second vehicle coordinates, respectively. Based on the first and second vehicle coordinates and the corresponding acquisition time, a mapping relationship between the coordinates of the boundary point of the first object and the image acquisition time is accurately determined. This mapping relationship can then be used to accurately determine the positional relationship between the first object and the vehicle's travel path at the moment of collision. This positional relationship can then be used to accurately control the vehicle's automatic emergency braking function, reducing the occurrence of false braking.

[0154] In some embodiments, since the vehicle coordinate system and the image coordinate system can be associated with each other based on the camera coordinate system of the image acquisition device, in the process of converting the image coordinates (first image coordinates and second image coordinates) into the vehicle coordinates (first vehicle coordinates and second vehicle coordinates) in S4041, the image coordinates can be first converted into the camera coordinates in the camera coordinate system, and then converted into the vehicle coordinates in the vehicle coordinate system. Figure 9 , refer to Figure 10 As shown, S4041 may specifically include S1001-S1004:

[0155] S1001. Determine a longitudinal distance between a first object and a vehicle at a first moment.

[0156] In the embodiment of the present disclosure, the longitudinal distance between the first object and the vehicle may refer to the longitudinal distance between the image acquisition device and the vehicle. The longitudinal distance may be obtained in any possible manner, and the present disclosure does not impose any specific limitation on this.

[0157] S1002 : Based on the first image coordinates, the longitudinal distance, and the focal length of the image acquisition device of the vehicle, convert the first image coordinates into first camera coordinates in a camera coordinate system.

[0158] In some embodiments, reference Figure 11 As shown, based on the perspective projection theorem, we know that the following formula exists between the image coordinate system and the camera coordinate system:

[0159]

[0160] Wherein w is the first pixel distance between two pixel points on the first object in the image, f is the focal length of the image acquisition device on the vehicle, W is the actual length corresponding to the first pixel distance on the first object, and Z is the longitudinal distance between the image acquisition device and the first object in the longitudinal direction.

[0161] In one possible implementation, if the origin of the image coordinate system is the projection of the origin in the camera coordinate system, then the X-axis coordinate of the left boundary point of the first object (e.g., the lower left boundary point) in the first camera coordinates and the X-axis coordinate of the left boundary point of the first object in the first image coordinates have the following conversion relationship:

[0162]

[0163] Among them, X l is the X-axis coordinate of the left boundary point of the first object in the first camera coordinate, x l is the X-axis coordinate of the left boundary point of the first object in the first image coordinates, f is the focal length of the image acquisition device on the vehicle, and Z0 is the longitudinal distance between the vehicle (specifically the image acquisition device of the vehicle) and the first object at the first moment.

[0164] Similarly, the Y-axis coordinate of the left boundary point of the first object in the first camera coordinates and the Y-axis coordinate of the left boundary point of the first object in the first image coordinates have the following conversion relationship:

[0165]

[0166] Among them, Y l is the Y-axis coordinate of the left boundary point of the first object in the first camera coordinate, y l is the Y-axis coordinate of the left boundary point of the first object in the first image coordinates.

[0167] In the above implementation, the conversion relationship between the first camera coordinates (X-axis and Y-axis coordinates) of the right boundary point of the first object and the first image coordinates of the right boundary point of the first object is similar. Furthermore, the Z-axis coordinates of the first camera coordinates of the boundary points (left and right boundary points) of the first object are Z0.

[0168] In another possible implementation, if the origin of the image coordinate system is not the projection of the origin in the camera coordinate system, then the X-axis coordinate of the left boundary point of the first object (e.g., the lower left boundary point) in the first camera coordinates and the X-axis coordinate of the left boundary point of the first object in the first image coordinates have the following conversion relationship:

[0169]

[0170] Among them, u0 is the X-axis coordinate of the vanishing point in the intrinsic parameters of the image acquisition device. The specific physical meaning of u0 is the X-axis coordinate of the projection point of the origin of the camera coordinate system in the image coordinate system.

[0171] Similarly, the Y-axis coordinate of the left boundary point of the first object in the first camera coordinates and the Y-axis coordinate of the left boundary point of the first object in the first image coordinates have the following conversion relationship:

[0172]

[0173] Among them, v0 is the Y-axis coordinate of the vanishing point in the internal parameters of the image acquisition device. The specific physical meaning of v0 is the Y-axis coordinate of the projection point of the origin of the camera coordinate in the image coordinate system.

[0174] In the above implementation, the conversion relationship between the first camera coordinates (X-axis and Y-axis coordinates) of the right boundary point of the first object and the first image coordinates of the right boundary point of the first object is similar. Furthermore, the Z-axis coordinates of the first camera coordinates of the boundary points (left and right boundary points) of the first object are Z0.

[0175] When the first camera coordinates are obtained, the first camera coordinates may be converted into the first vehicle coordinates based on the transformation relationship between the camera coordinate system and the vehicle coordinate system, that is, S1004 is executed.

[0176] S1003 : Based on the second image coordinates, the longitudinal distance, and the focal length, convert the second image coordinates into second camera coordinates in a camera coordinate system.

[0177] The specific implementation of S1003 can refer to the specific implementation of S1002, which will not be repeated here.

[0178] When the second camera coordinates are obtained, the second camera coordinates may be converted into the first vehicle coordinates based on the transformation relationship between the camera coordinate system and the vehicle coordinate system, that is, S1004 is executed.

[0179] It should be noted that there is no necessary order between S1002 and S1003. Figure 10 The examples shown are only examples. In practice, S1002 may be executed first, or S1003 may be executed first, or S1002 and S1003 may be executed simultaneously. This disclosure does not impose any specific restrictions on this.

[0180] S1004 : Based on the transformation relationship between the camera coordinate system and the ego-vehicle coordinate system, the first camera coordinates are converted into the first ego-vehicle coordinates, and the second camera coordinates are converted into the second ego-vehicle coordinates.

[0181] The technical solution provided by the embodiments of the present disclosure utilizes the conversion relationship between the image coordinate system and the camera coordinate system to accurately obtain first and second camera coordinates based on the longitudinal distance between the first object and the vehicle at a first moment and the focal length of the image acquisition device. Furthermore, based on the conversion relationship between the camera coordinate system and the vehicle coordinate system, accurate first and second ego-vehicle coordinates are obtained. Furthermore, based on the first and second ego-vehicle coordinates and their corresponding acquisition moments, an accurate mapping relationship and positional relationship can be subsequently determined. This positional relationship can then be used to accurately control the automatic emergency braking function, minimizing the occurrence of false braking.

[0182] Since the longitudinal distance between the first object and the vehicle may change during the driving process, Figure 11According to the perspective projection theorem shown in the figure, when converting the second image coordinates to the second camera coordinates, the longitudinal distance between the vehicle and the first object at the second moment needs to be used, rather than the longitudinal distance between the vehicle and the first object at the first moment. However, since the calculation of the longitudinal distance itself has a certain error, if the longitudinal distance between the vehicle and the first object at the second moment is used to calculate the second camera coordinates, and then the mapping relationship is determined based on the second vehicle coordinates converted from the second camera coordinates, this error will be accumulated in the mapping relationship. At the same time, since the first object is a rigid body, its actual width does not change. Based on the principle of image capture that objects appear larger when they are closer and smaller when they are farther away, the width of the first object in the image is in a fixed inverse relationship with the longitudinal distance, which can be obtained by the following formula:

[0183] w0×Z0=w i ×Z i (18)

[0184] Wherein, w0 is the distance between the boundary points of the first object in the first image, that is, the width of the first object in the first image; w i is the distance between the boundary points of the first object in the second image, that is, the width of the first object in the second image; Z i is the longitudinal distance between the vehicle and the first object at the second moment when the second image is captured.

[0185] Based on this, in some embodiments, in order to obtain accurate second camera coordinates, steps S1-S4 are further included before S1004:

[0186] S1. Determine a first boundary width of a first object in a first image based on first image coordinates.

[0187] The first boundary width is the distance between the first image coordinates of the left boundary point and the first image coordinates of the right boundary point of the first object.

[0188] S2. Determine a second boundary width of the first object in the second image based on the second image coordinates.

[0189] The second boundary width is the distance between the second image coordinates of the left boundary point and the second image coordinates of the right boundary point of the first object.

[0190] S3. Determine a distance correction factor of the first object in the second image based on the first boundary width and the second boundary width.

[0191] For example, the calculation formula of the distance correction factor can be obtained by combining the above formula (18) as follows:

[0192]

[0193] Among them, w scale is the distance correction factor.

[0194] It should be noted that S1-S3 can be executed at any possible time before S4, and the present disclosure does not impose any specific limitation on this.

[0195] S4. Correct the second camera coordinates based on the distance correction factor of the first object in the second image.

[0196] In one possible implementation, combining the above formulas (14) and (15) yields:

[0197]

[0198] Among them, X li is the X-axis coordinate of the left boundary point of the first object in the second camera coordinates, x li is the X-axis coordinate of the left boundary point of the first object in the second image coordinates, Y li is the Y-axis coordinate of the left boundary point of the first object in the second camera coordinates, y li is the Y-axis coordinate of the left boundary point of the first object in the second image coordinates.

[0199] In the above implementation, the conversion relationship between the second camera coordinates (X-axis and Y-axis coordinates) of the right boundary point of the first object and the second image coordinates of the right boundary point of the first object is similar. In addition, the Z-axis coordinates of the boundary points (left and right boundary points) of the first object in the second camera coordinates are Z0.

[0200] In another possible implementation, combining the above formulas (16) and (17) yields:

[0201]

[0202] In the above implementation, the conversion relationship between the second camera coordinates (X-axis coordinates and Y-axis coordinates) of the right boundary point of the first object and the second image coordinates of the right boundary point of the first object is the same. In addition, the Z-axis coordinates of the second camera coordinates of the boundary points (left boundary points and right boundary points) of the first object are Z0×w scale .

[0203] The technical solution provided by the embodiment of the present disclosure converts the longitudinal distance between the vehicle and the first object at the first moment based on the principle that the width of the first object in the image is inversely proportional to the longitudinal distance, thereby completing the correction of the second camera coordinates and obtaining more accurate second camera coordinates. Subsequently, more accurate mapping relationships and positional relationships can be obtained, and the automatic emergency braking function can be accurately controlled based on the positional relationship, thereby reducing the occurrence of false braking.

[0204] In some embodiments, combined Figure 4 , refer to Figure 12As shown, S407 may specifically include S4071 and S4072:

[0205] S4071: In response to the position relationship that the first object is in the driving path of the vehicle at the moment of collision, triggering of the automatic emergency braking function is not suppressed.

[0206] At the moment of collision, if the first object is in the vehicle's travel path, there is a risk of collision between the vehicle and the first object, and in this case, there is no need to suppress the triggering of the automatic emergency braking function. Whether the automatic emergency braking function is subsequently triggered may be determined by a subsequent algorithm in the vehicle's control device, and this disclosure does not impose specific limitations on this.

[0207] S4072: In response to the position relationship that the first object is not in the driving path of the vehicle at the moment of collision, suppressing the triggering of the automatic emergency braking function.

[0208] If the first object is in the vehicle's path at the moment of collision, it indicates that there is unlikely to be a risk of collision between the vehicle and the first object at the moment of collision, so it is necessary to suppress the triggering of the automatic emergency braking function to prevent accidental braking and ensure the driving safety of the vehicle.

[0209] It should be noted that, during one execution of the automatic emergency braking control method provided by the embodiment of the present disclosure, S4071 and S4072 are not executed simultaneously, but are executed one by one.

[0210] The technical solutions corresponding to S4071 and S4072 in this disclosure determine whether to suppress the triggering of the automatic emergency braking function based on the different positional relationships. This demonstrates that the entire solution's control of the automatic emergency braking function fully considers whether the first object is in the vehicle's travel path at the moment of collision. This results in more accurate automatic emergency braking control and reduces the occurrence of false braking.

[0211] Exemplary devices

[0212] Figure 13 FIG. 1 is a schematic diagram of the structure of an automatic emergency braking control device provided by an exemplary embodiment of the present disclosure. Figure 13 As shown, the automatic emergency braking control device may include:

[0213] A parameter determination module 1301 is configured to determine a rigid body transformation parameter of a first object between two adjacent frames of images based on an image sequence collected by the vehicle;

[0214] An acquisition module 1302 is configured to acquire first image coordinates of a boundary point of a first object in a first image captured by a vehicle at a first moment in an image coordinate system;

[0215] a boundary module 1303 configured to determine, based on the rigid body transformation parameters determined by the parameter determination module 1301 and the first image coordinates obtained by the acquisition module 1302, second image coordinates of a boundary point of the first object in the second image of the image sequence in the image coordinate system; the acquisition time of the second image is the second moment, which is earlier than the first moment;

[0216] A mapping module 1304 is configured to determine a mapping relationship between the coordinates of the boundary point of the first object in the ego-vehicle coordinate system and the image sequence acquisition time based on the first image coordinates determined by the parameter determination module 1301, the second image coordinates determined by the boundary module 1303, the first time, and the second time.

[0217] A collision module 1305 is configured to determine a collision time between the vehicle and the first object at a first moment;

[0218] a position determination module 1306 for determining a positional relationship between the first object and the vehicle's travel path based on the object collision time determined by the collision module 1305 and the mapping relationship determined by the mapping module 1304;

[0219] The processing module 1307 is configured to control the automatic emergency braking function of the vehicle based on the position relationship determined by the position determination module 1306 .

[0220] In some embodiments, the parameter determination module 1301 is specifically configured to perform optical flow tracking on a first object in an image sequence, and determine rigid body transformation parameters of the first object between two adjacent frames of images.

[0221] In some embodiments, the acquisition module 1302 is specifically configured to perform optical flow tracking on a first object in a third image in the image sequence that is captured earlier than the first image, to obtain first image coordinates.

[0222] In some embodiments, combined Figure 13 , refer to Figure 14 As shown, the mapping module 1304 may include a conversion unit 1401 and a mapping unit 1402. The conversion unit 1401 is configured to convert the first image coordinates into first ego-vehicle coordinates in an ego-vehicle coordinate system, and convert the second image coordinates into second ego-vehicle coordinates in an ego-vehicle coordinate system; and the mapping unit 1402 is configured to determine a mapping relationship between the coordinates of a boundary point of the first object in the ego-vehicle coordinate system and the capture time of an image in the image sequence based on the first ego-vehicle coordinates converted by the conversion unit 1401, the second ego-vehicle coordinates converted by the conversion unit 1401, the first moment, and the second moment.

[0223] In some embodiments, the conversion unit 1401 includes a first subunit, a second subunit, and a third subunit. The first subunit is configured to determine the longitudinal distance between the first object and the vehicle at a first moment; the second subunit is configured to convert the first image coordinates into first camera coordinates in a camera coordinate system based on the first image coordinates, the longitudinal distance, and the focal length of the vehicle's image acquisition device; the second subunit is further configured to convert the second image coordinates into second camera coordinates in the camera coordinate system based on the second image coordinates, the longitudinal distance, and the focal length; and the third subunit is configured to convert the first camera coordinates into first ego-vehicle coordinates and the second camera coordinates into second ego-vehicle coordinates based on a transformation relationship between the camera coordinate system and the ego-vehicle coordinate system.

[0224] In some embodiments, the conversion unit 1401 further includes a fourth subunit, a fifth subunit, and a sixth subunit. The fourth subunit is configured to determine a first border width of the first object in the first image based on the first image coordinates; the fourth subunit is further configured to determine a second border width of the first object in the second image based on the second image coordinates; the fifth subunit is configured to determine a distance correction factor of the first object in the second image based on the first border width and the second border width determined by the fourth subunit; and the sixth subunit is configured to correct the second camera coordinates based on the distance correction factor of the first object in the second image determined by the fifth subunit.

[0225] In some embodiments, reference Figure 14 As shown, the position determination module 1306 specifically includes a coordinate determination unit 1403 and a relationship determination unit 1404. The coordinate determination unit 1403 is configured to determine the coordinates of the boundary point of the first object in the ego-vehicle coordinate system at the moment of collision based on the mapping relationship and the object collision time; and the relationship determination unit 1404 is configured to determine the positional relationship between the first object and the vehicle's travel path at the moment of collision based on the coordinates of the boundary point of the vehicle in the ego-vehicle coordinate system and the coordinates of the boundary point of the first object in the ego-vehicle coordinate system at the moment of collision determined by the coordinate determination unit 1403.

[0226] In some embodiments, the relationship determination unit 1404 specifically includes a range subunit, a first processing unit, and a second processing unit. The range subunit is configured to determine the boundary range of the vehicle in the ego-vehicle coordinate system at the time of collision based on the coordinates of the vehicle's boundary points in the ego-vehicle coordinate system. The first processing unit is configured to, in response to the coordinates of the object's boundary points being within the boundary range, determine that the positional relationship indicates that the first object was in the vehicle's travel path at the time of collision. The second processing unit is configured to, in response to the coordinates of the object's boundary points being outside the boundary range, determine that the positional relationship indicates that the first object was not in the vehicle's travel path at the time of collision.

[0227] In some embodiments, combined Figure 13 , refer to Figure 14As shown, the processing module 1307 includes a non-inhibition unit 1405 and a suppression unit 1406. The non-inhibition unit 1405 is configured to not suppress the triggering of the automatic emergency braking function in response to the positional relationship that the first object is in the vehicle's driving path at the time of collision; and the suppression unit 1406 is configured to suppress the triggering of the automatic emergency braking function in response to the positional relationship that the first object is not in the vehicle's driving path at the time of collision.

[0228] In some embodiments, the collision module 1305 is specifically used to: perform optical flow tracking on the first object in the third image in the image sequence whose acquisition time is earlier than the first image, obtain the image domain collision time between the vehicle and the first object at the first moment, and determine the image domain collision time as the object collision time.

[0229] In some embodiments, the collision module 1305 is further configured to, in response to an abnormality in the image-domain collision time, determine the numerical-domain collision time between the vehicle and the first object at the first moment as the object collision time.

[0230] It should be noted that the description of the above exemplary embodiment of the device is similar to the description of the above exemplary embodiment of the method, and has the same beneficial effects as the corresponding exemplary embodiment of the method. For technical details and corresponding beneficial technical effects not disclosed in the exemplary embodiment of the device of the present disclosure, those skilled in the art should refer to the description of the exemplary embodiment of the method of the present disclosure for understanding, and will not be repeated here.

[0231] Exemplary electronic devices

[0232] Figure 15 A structural diagram of an electronic device provided in an embodiment of the present disclosure includes at least one processor 151 and a memory 152 for storing processor-executable instructions.

[0233] The processor 151 may be a central processing unit (CPU) or other forms of processing units having data processing capabilities and / or instruction execution capabilities, and may control other components in the electronic device to perform desired functions.

[0234] The memory 152 may include one or more computer program products, which may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and / or cache memory. Non-volatile memory may include, for example, read-only memory (ROM), a hard disk, flash memory, etc. One or more computer program instructions may be stored on the computer-readable storage medium, and the processor 151 may execute the one or more computer program instructions to implement the automatic emergency braking control method and / or other desired functions of the various embodiments of the present disclosure described above.

[0235] In one example, the electronic device may further include an input device 153 and an output device 154 , and these components are interconnected via a bus system and / or other forms of connection mechanisms (not shown).

[0236] The input device 153 may also include, for example, a keyboard, a mouse, a touch screen, etc.

[0237] The output device 154 can output various information to the outside, and may include, for example, a display, a speaker, a printer, a communication network and its connected remote output devices, etc.

[0238] Of course, to simplify, Figure 15 Only some of the components related to the present disclosure in the electronic device are shown, and components such as a bus, an input / output interface, etc. are omitted. In addition, the electronic device may further include any other appropriate components according to specific application scenarios.

[0239] Exemplary computer program products and computer-readable storage media

[0240] In addition to the above-mentioned methods and devices, embodiments of the present disclosure may also provide a computer program product, including computer program instructions, which, when executed by a processor, enable the processor to execute the steps of the automatic emergency braking control method of various embodiments of the present disclosure described in the above-mentioned "Exemplary Method" section.

[0241] The computer program product may be written in any combination of one or more programming languages ​​to implement the operations of the disclosed embodiments, including object-oriented programming languages ​​such as Java, C++, and conventional procedural programming languages ​​such as C or similar programming languages. The program code may be executed entirely on the user's computing device, partially on the user's computing device, as a stand-alone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0242] In addition, an embodiment of the present disclosure may also be a computer-readable storage medium having computer program instructions stored thereon, which, when executed by a processor, enables the processor to execute the steps of the automatic emergency braking control method of various embodiments of the present disclosure described in the above-mentioned "Exemplary Method" section.

[0243] Computer readable storage media can adopt any combination of one or more readable media. The readable medium can be a readable signal medium or a readable storage medium. The readable storage medium is, for example, but not limited to, a system, device or component comprising electricity, magnetism, light, electromagnetic, infrared, or semiconductor, or any combination thereof. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or 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 thereof.

[0244] The basic principles of the present disclosure have been described above in conjunction with specific embodiments. However, the advantages, strengths, and effects mentioned in this disclosure are merely illustrative and not restrictive, and should not be considered as essential to each embodiment of the present disclosure. Furthermore, the specific details disclosed above are provided for illustrative purposes and to facilitate understanding, rather than as limitations. These details do not limit the present disclosure to necessarily being implemented using these specific details.

[0245] Those skilled in the art may make various changes and modifications to the present disclosure without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present disclosure and their equivalents, the present disclosure is intended to include these modifications and variations.

Claims

1. An automatic emergency braking control method, comprising: Determining, based on a sequence of images captured by the vehicle, rigid body transformation parameters of a first object between two adjacent frames of images; Obtaining first image coordinates of a boundary point of a first object in a first image captured by the vehicle at a first moment in an image coordinate system; determining, based on the rigid body transformation parameters and the first image coordinates, second image coordinates of a boundary point of a first object in a second image in the image sequence in an image coordinate system; the acquisition time of the second image is a second moment, and the second moment is earlier than the first moment; Determining a mapping relationship between the coordinates of the boundary point of the first object and the capture time of the images in the image sequence in the vehicle coordinate system based on the first image coordinates, the second image coordinates, the first time, and the second time; determining a collision time between the vehicle and the first object at the first moment; Determining a positional relationship between the first object and the vehicle's travel path at the moment of collision based on the object collision time and the mapping relationship; Based on the positional relationship, an automatic emergency braking function of the vehicle is controlled.

2. The method according to claim 1, wherein The determining of rigid body transformation parameters of the first object between two adjacent frames of images based on the image sequence collected by the vehicle includes: Optical flow tracking is performed on a first object in the image sequence to determine rigid body transformation parameters of the first object between two adjacent frames of images.

3. The method according to claim 1, wherein The obtaining of first image coordinates of a boundary point of a first object in a first image captured by the vehicle at a first moment in an image coordinate system includes: Optical flow tracking is performed on a first object in a third image in the image sequence that is acquired earlier than the first image to obtain the first image coordinates.

4. The method according to claim 1, wherein The determining, based on the first image coordinates, the second image coordinates, the first moment, and the second moment, a mapping relationship between the coordinates of the boundary point of the first object in the vehicle coordinate system and the acquisition moments of the images in the image sequence includes: Converting the first image coordinates into first ego-vehicle coordinates in an ego-vehicle coordinate system, and converting the second image coordinates into second ego-vehicle coordinates in an ego-vehicle coordinate system; Based on the first ego-vehicle coordinates, the second ego-vehicle coordinates, the first moment, and the second moment, a mapping relationship between the coordinates of the boundary points of the first object in the ego-vehicle coordinate system and the capture moments of the images in the image sequence is determined.

5. The method according to claim 4, wherein The converting the first image coordinates into first ego-vehicle coordinates in the ego-vehicle coordinate system, and converting the second image coordinates into second ego-vehicle coordinates in the ego-vehicle coordinate system, includes: determining a longitudinal distance between the first object and the vehicle at the first moment; Converting the first image coordinates into first camera coordinates in a camera coordinate system based on the first image coordinates, the longitudinal distance, and a focal length of an image acquisition device of the vehicle; converting the second image coordinates into second camera coordinates in a camera coordinate system based on the second image coordinates, the longitudinal distance, and the focal length; Based on a transformation relationship between a camera coordinate system and a vehicle coordinate system, the first camera coordinates are transformed into the first vehicle coordinates, and the second camera coordinates are transformed into the second vehicle coordinates.

6. The method according to claim 5, wherein: The converting the second image coordinates into second vehicle coordinates in the vehicle coordinate system further includes: determining a first boundary width of a first object in the first image based on the first image coordinates; determining a second boundary width of the first object in the second image based on the second image coordinates; determining a distance correction factor of a first object in the second image based on the first boundary width and the second boundary width; The second camera coordinates are corrected based on a distance correction factor of the first object in the second image.

7. The method according to claim 1, wherein The determining, based on the collision time and the mapping relationship, a positional relationship between the first object and the vehicle's travel path at the collision moment includes: Determining the coordinates of the object boundary point of the first object in the vehicle coordinate system at the moment of collision based on the mapping relationship and the object collision time; Based on the coordinates of the boundary point of the vehicle in the vehicle coordinate system and the coordinates of the object boundary point of the first object in the vehicle coordinate system at the moment of collision, a positional relationship between the first object and the vehicle driving path at the moment of collision is determined.

8. The method according to claim 7, wherein: The determining, based on the coordinates of the boundary point of the vehicle in the vehicle coordinate system and the coordinates of the boundary point of the first object in the vehicle coordinate system at the time of collision, a positional relationship between the first object and the vehicle's travel path at the time of collision includes: Determining a boundary range of the vehicle in the ego-vehicle coordinate system at the moment of collision based on coordinates of the boundary points of the vehicle in the ego-vehicle coordinate system; In response to the coordinates of the boundary point of the object being within the boundary range, determining the positional relationship as the first object being in the driving path of the vehicle at the moment of collision; In response to the coordinates of the boundary point of the object being outside the boundary range, the positional relationship is determined to be that the first object is not in the driving path of the vehicle at the moment of collision.

9. The method according to claim 8, wherein Based on the positional relationship, controlling an automatic emergency braking function of the vehicle includes: In response to the positional relationship being that the first object is in the travel path of the vehicle at the moment of the collision, triggering of an automatic emergency braking function is not suppressed; In response to the positional relationship being that the first object is not in the travel path of the vehicle at the moment of the collision, triggering of an automatic emergency braking function is suppressed.

10. The method according to claim 1, wherein Determining the collision time between the vehicle and the first object at the first moment includes: Optical flow tracking is performed on a first object in a third image in the image sequence that is captured earlier than the first image to obtain an image-domain collision time between the vehicle and the first object at the first moment, and the image-domain collision time is determined as the object collision time.

11. The method according to claim 10, wherein: The determining of the collision time between the vehicle and the first object at the first moment further includes: In response to the abnormality in the image-domain collision time, a value-domain collision time between the vehicle and the first object at the first moment is determined as the object collision time.

12. An automatic emergency braking control device comprising: a parameter determination module, configured to determine a rigid body transformation parameter of a first object between two adjacent frames of images based on a sequence of images acquired by the vehicle; An acquisition module, configured to acquire a first image coordinate of a boundary point of a first object in a first image captured by the vehicle at a first moment in an image coordinate system; a boundary module, configured to determine, based on the rigid body transformation parameters determined by the parameter determination module and the first image coordinates acquired by the acquisition module, second image coordinates of a boundary point of the first object in a second image in the image sequence in an image coordinate system; wherein the acquisition time of the second image is a second moment, and the second moment is earlier than the first moment; a mapping module, configured to determine a mapping relationship between the coordinates of the boundary points of the first object in the ego-vehicle coordinate system and the image sequence acquisition time based on the first image coordinates determined by the parameter determination module, the second image coordinates determined by the boundary module, the first time, and the second time; a collision module, configured to determine a collision time between the vehicle and the first object at the first moment; a position determination module, configured to determine a positional relationship between a first object and the vehicle's travel path based on the object collision time determined by the collision module and the mapping relationship determined by the mapping module; A processing module is used to control the automatic emergency braking function of the vehicle based on the position relationship determined by the position determination module.

13. A computer-readable storage medium storing a computer program, wherein the computer program is used to execute the automatic emergency braking control method according to any one of claims 1 to 11.

14. An electronic device, comprising: processor; a memory for storing instructions executable by the processor; The processor is used to read the executable instructions from the memory and execute the instructions to implement the automatic emergency braking control method described in any one of claims 1-11 above.

Images