Emergency braking triggering method, device, electronic device and vehicle

Abstract

The application provides an emergency braking triggering method and device, electronic equipment and vehicle. The method comprises the following steps: obtaining initial attribute information of a plurality of initial targets detected by a millimeter wave radar and a camera; obtaining a plurality of fusion targets and corresponding fusion attribute information of each fusion target according to the initial attribute information of the plurality of initial targets; removing false targets from the plurality of fusion targets to obtain real fusion targets; and calculating the fusion collision time, the first initial collision time and the second initial collision time of each real fusion target according to the fusion attribute information of each real fusion target. When the collision time of the fusion target is less than a preset collision time, and the difference between the first initial collision time and the second initial collision time is less than a preset collision time difference, the emergency braking selection result is determined as automatic emergency braking, so that false fusion targets can be removed, and AEB false triggering caused by false fusion targets can be avoided.

Description

Technical Field

[0001] This invention relates to the field of vehicle technology, and more particularly to an emergency braking triggering method, device, electronic equipment, and vehicle. Background Technology

[0002] The Advanced Emergency Braking System (AEBS) uses a forward-facing camera and a forward-facing medium-to-long-range millimeter-wave radar to detect vehicles, pedestrians, and obstacles ahead, and automatically activates the vehicle's braking system to slow the vehicle down when a collision is possible, in order to avoid a collision or mitigate the consequences of a collision.

[0003] In related technologies, due to the multipath effect of millimeter-wave radar and the susceptibility of cameras to weather and lighting conditions, false targets are easily generated simultaneously, thus easily leading to false fused targets. If the collision time of the false fused targets meets the triggering conditions of AEB (Autonomous Emergency Braking), it will result in false AEB triggering. Summary of the Invention

[0004] This invention provides an emergency braking triggering method, device, electronic device, and vehicle, which aims to remove false fusion targets by fusing target constraints, thereby avoiding false AEB triggering caused by false fusion targets.

[0005] To solve the above-mentioned technical problems, the present invention is implemented as follows:

[0006] In a first aspect, embodiments of this application provide an emergency braking triggering method, including:

[0007] Acquire initial attribute information of multiple initial targets detected by millimeter-wave radar and cameras;

[0008] Based on the initial attribute information of the multiple initial targets, target fusion is performed on the multiple initial targets detected by the millimeter-wave radar and the multiple initial targets detected by the camera to obtain multiple fused targets and fused attribute information corresponding to each fused target. The fused attribute information includes at least lateral distance, longitudinal distance, lateral velocity, longitudinal velocity and fused target survival time.

[0009] Based on the fusion target constraints and the fusion attribute information corresponding to each fusion target, false targets are eliminated from the multiple fusion targets to obtain the true fusion targets;

[0010] Based on the fusion attribute information of each real fusion target, the fusion collision time, the first initial collision time and the second initial collision time of each real fusion target are calculated.

[0011] When the collision time of the real fusion target is less than the preset collision time, and the difference between the first initial collision time and the second initial collision time is less than the preset collision time difference, the emergency braking selection result is determined to be automatic emergency braking.

[0012] Optionally, the fusion target constraint includes at least one of the following:

[0013] The difference between the lateral distances detected by the millimeter-wave radar and the camera is less than a preset lateral distance difference;

[0014] The survival time of the fusion target is greater than the first preset survival time;

[0015] The difference between the longitudinal distances detected by the millimeter-wave radar and the camera is less than a preset longitudinal distance difference;

[0016] The difference between the longitudinal velocity detected by the millimeter-wave radar and the camera is less than a preset longitudinal velocity difference;

[0017] When the fusion target is a vehicle, the survival time of the fusion target is greater than the second preset survival time; when the fusion target is a pedestrian, the survival time of the fusion target is greater than the third preset survival time, and the second preset survival time is greater than the third preset survival time.

[0018] Optionally, based on the initial attribute information of the multiple initial targets, target fusion is performed on the multiple initial targets detected by the millimeter-wave radar and the multiple initial targets detected by the camera to obtain multiple fused targets and fused attribute information corresponding to each fused target, including:

[0019] The initial attribute information of multiple initial targets detected by the millimeter-wave radar and camera is parameterized to obtain the updated attribute information of multiple initial targets detected by the millimeter-wave radar and camera.

[0020] Based on the updated attribute information of the multiple initial targets, the multiple initial targets detected by the millimeter-wave radar are associated with the multiple initial targets detected by the camera to obtain the optimal association result between the multiple selected targets detected by the millimeter-wave radar and the multiple selected targets detected by the camera.

[0021] Based on the optimal association result, target fusion is performed on the multiple initial targets detected by the millimeter-wave radar and the multiple initial targets detected by the camera that are related to each other, to obtain multiple fused targets and fused attribute information corresponding to each fused target.

[0022] Optionally, the first initial collision time and the second initial collision time of the real fusion target are obtained by the following method:

[0023] The updated attribute information detected by millimeter-wave radar and camera is obtained from the fusion attribute information of the real fusion target;

[0024] The first initial collision time is obtained based on the updated attribute information detected by the millimeter-wave radar of the real fused target;

[0025] The second initial collision time is obtained based on the updated attribute information detected by the camera of the real fusion target.

[0026] Optionally, the updated attribute information includes the distance from the initial target to the vehicle, the standard deviation of the millimeter-wave radar velocity, and the number of stable tracking frames by the camera;

[0027] Based on the updated attribute information of the multiple initial targets, the multiple initial targets detected by the millimeter-wave radar are associated with the multiple initial targets detected by the camera to obtain the optimal association result between the multiple selected targets detected by the millimeter-wave radar and the multiple selected targets detected by the camera, including:

[0028] The camera noise intensity value and the millimeter-wave radar noise intensity value are determined based on the distance from the initial target to the vehicle, the standard deviation of the millimeter-wave radar velocity, and the number of stable tracking frames of the camera.

[0029] Based on the camera noise intensity value and the millimeter-wave radar noise intensity value, the fusion weights of multiple initial targets detected by the millimeter-wave radar and the fusion weights of multiple initial targets detected by the camera are determined. The fusion weights of the multiple initial targets detected by the camera are negatively correlated with the camera noise intensity value, and the fusion weights of the multiple initial targets detected by the millimeter-wave radar are negatively correlated with the millimeter-wave radar noise intensity value.

[0030] Based on the fusion weights of the multiple initial targets detected by the millimeter-wave radar and the multiple initial targets detected by the camera, the optimal correlation result between the multiple initial targets detected by the millimeter-wave radar and the multiple initial targets detected by the camera is obtained.

[0031] Optionally, before the step of obtaining the optimal correlation result between the multiple initial targets detected by the millimeter-wave radar and the multiple initial targets detected by the camera based on the fusion weights of the multiple initial targets detected by the millimeter-wave radar and the fusion weights of the multiple initial targets detected by the camera, the method further includes:

[0032] For each initial target among the multiple screened targets detected by the millimeter-wave radar, if the initial target is a single radar target, and the lateral distance of the initial target is less than a preset lateral distance, and the absolute value of the lateral velocity of the initial target is greater than a preset absolute value of the lateral velocity, and the initial target is in a recursive state, and the number of detection frames of the initial target is less than a preset number of frames, the initial target is eliminated.

[0033] Optionally, the fusion attribute information of the fusion target is obtained through the following method:

[0034] The lateral distance detected by the camera is determined as the fusion lateral distance of the fusion target;

[0035] The longitudinal distance detected by the millimeter-wave radar is determined as the fused longitudinal distance of the fused target;

[0036] The lateral velocity detected by the camera is determined as the fusion lateral velocity of the fusion target;

[0037] The longitudinal velocity detected by the millimeter-wave radar is determined as the fused longitudinal velocity of the fused target.

[0038] Secondly, embodiments of the present invention provide an emergency braking triggering device, comprising:

[0039] The acquisition module is used to acquire initial attribute information of multiple initial targets detected by millimeter-wave radar and cameras;

[0040] The fusion module is used to perform target fusion on multiple initial targets detected by the millimeter-wave radar and multiple initial targets detected by the camera based on the initial attribute information of the multiple initial targets, to obtain multiple fused targets and fused attribute information corresponding to each fused target. The fused attribute information includes at least lateral distance, longitudinal distance, lateral velocity, longitudinal velocity and fused target survival time.

[0041] The elimination module is used to eliminate false targets from the multiple fusion targets based on the fusion target constraints and the fusion attribute information corresponding to each fusion target, so as to obtain the true fusion targets;

[0042] The calculation module is used to calculate the fusion collision time, the first initial collision time and the second initial collision time of each real fusion target based on the fusion attribute information of each real fusion target.

[0043] The determination module is used to determine the emergency braking selection result as automatic emergency braking when the collision time of the real fusion target is less than a preset collision time, and the difference between the first initial collision time and the second initial collision time is less than a preset collision time difference.

[0044] Optionally, the fusion target constraint includes at least one of the following:

[0045] The difference between the lateral distances detected by the millimeter-wave radar and the camera is less than a preset lateral distance difference;

[0046] The survival time of the fusion target is greater than the first preset survival time;

[0047] The difference between the longitudinal distances detected by the millimeter-wave radar and the camera is less than a preset longitudinal distance difference;

[0048] The difference between the longitudinal velocity detected by the millimeter-wave radar and the camera is less than a preset longitudinal velocity difference;

[0049] When the fusion target is a vehicle, the survival time of the fusion target is greater than the second preset survival time; when the fusion target is a pedestrian, the survival time of the fusion target is greater than the third preset survival time, and the second preset survival time is greater than the third preset survival time.

[0050] Optionally, the fusion module includes:

[0051] The unification submodule is used to unify the initial attribute information of multiple initial targets detected by the millimeter-wave radar and camera, and obtain the updated attribute information of multiple initial targets detected by the millimeter-wave radar and camera.

[0052] The association submodule is used to associate the multiple initial targets detected by the millimeter-wave radar with the multiple initial targets detected by the camera based on the updated attribute information of the multiple initial targets, so as to obtain the optimal association result between the multiple selected targets detected by the millimeter-wave radar and the multiple selected targets detected by the camera;

[0053] The fusion submodule is used to perform target fusion on the multiple initial targets detected by the millimeter-wave radar and the multiple initial targets detected by the camera that are related to each other, based on the optimal association result, so as to obtain multiple fused targets and fusion attribute information corresponding to each fused target.

[0054] Optionally, the computing module includes:

[0055] The acquisition submodule is used to acquire the updated attribute information detected by millimeter-wave radar and camera contained in the fusion attribute information of the real fusion target;

[0056] The first acquisition submodule is used to obtain the first initial collision time based on the updated attribute information detected by the millimeter-wave radar of the real fused target.

[0057] The second acquisition submodule is used to obtain the second initial collision time based on the updated attribute information detected by the camera of the real fusion target.

[0058] Optionally, the updated attribute information includes the distance from the initial target to the vehicle, the standard deviation of the millimeter-wave radar velocity, and the number of stable tracking frames by the camera;

[0059] The associated sub-modules include:

[0060] The first determining subunit is used to determine the camera noise intensity value and the millimeter-wave radar noise intensity value based on the distance from the initial target to the vehicle, the standard deviation of the millimeter-wave radar velocity, and the number of stable tracking frames of the camera.

[0061] The second determining subunit is used to determine the fusion weights of multiple initial targets detected by the millimeter-wave radar and the fusion weights of multiple initial targets detected by the camera based on the camera noise intensity value and the millimeter-wave radar noise intensity value. The fusion weights of the multiple initial targets detected by the camera are negatively correlated with the camera noise intensity value, and the fusion weights of the multiple initial targets detected by the millimeter-wave radar are negatively correlated with the millimeter-wave radar noise intensity value.

[0062] The association subunit is used to obtain the optimal association result between the multiple initial targets detected by the millimeter-wave radar and the multiple initial targets detected by the camera, based on the fusion weights of the multiple initial targets detected by the millimeter-wave radar and the fusion weights of the multiple initial targets detected by the camera.

[0063] Optionally, prior to the associated subunit, the device further includes:

[0064] The elimination subunit is used to eliminate each initial target among multiple initial targets detected by the millimeter-wave radar when the initial target is a single radar target, the lateral distance of the initial target is less than a preset lateral distance, the absolute value of the lateral velocity of the initial target is greater than a preset absolute value of the lateral velocity, the initial target is in a recursive state, and the number of detection frames of the initial target is less than a preset number of frames.

[0065] Optionally, the fusion module includes:

[0066] The first determining submodule is used to determine the lateral distance detected by the camera as the fusion lateral distance of the fusion target;

[0067] The first determining submodule is used to determine the longitudinal distance detected by the millimeter-wave radar as the fused longitudinal distance of the fused target;

[0068] The first determining submodule is used to determine the lateral velocity detected by the camera as the fusion lateral velocity of the fusion target;

[0069] The first determining submodule is used to determine the longitudinal velocity detected by the millimeter-wave radar as the fused longitudinal velocity of the fused target.

[0070] Thirdly, embodiments of the present invention also provide an electronic device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the computer program is executed by the processor, it implements the steps of the emergency braking triggering method described in the first aspect.

[0071] Fourthly, embodiments of the present invention further provide a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the emergency braking triggering method described in the first aspect.

[0072] Fifthly, embodiments of the present invention also provide a vehicle, including a vehicle body and an emergency braking triggering device disposed on the vehicle body, the emergency braking triggering device being used to perform the steps of the emergency braking triggering method described in the first aspect above.

[0073] In this invention, initial attribute information of multiple initial targets detected by millimeter-wave radar and camera is acquired. Based on this initial attribute information, target fusion is performed on the multiple initial targets detected by millimeter-wave radar and camera to obtain multiple fused targets and their corresponding fused attribute information. The fused attribute information includes at least lateral distance, longitudinal distance, lateral velocity, longitudinal velocity, and fused target survival time. Then, based on the fused target constraints and the fused attribute information of each fused target, false targets are eliminated to obtain true fused targets. Finally, based on the fused attribute information of each true fused target, the fusion collision time of each true fused target is calculated. The system considers the time between the first and second initial collisions. Finally, when the collision time of the fused target is less than the preset collision time, and the difference between the first and second initial collision times is less than the preset collision time difference, the emergency braking selection result is determined to be automatic emergency braking. After obtaining multiple fused targets, false fused targets are removed by fused target constraints. Only when the collision time of the real fused target is less than the preset collision time, and the difference between the first and second initial collision times is less than the preset collision time difference, is the emergency braking selection result determined to be automatic emergency braking. This further removes the interference of false fused targets, thereby avoiding false AEB triggering caused by false fused targets meeting the AEB triggering conditions. Attached Figure Description

[0074] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0075] Figure 1 This is a flowchart of the steps of an emergency braking triggering method according to an embodiment of the present invention;

[0076] Figure 2 This is a schematic diagram of an emergency braking triggering device according to an embodiment of the present invention;

[0077] Figure 3 This is a schematic diagram of the structure of an electronic device according to an embodiment of the present invention. Detailed Implementation

[0078] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0079] Automatic emergency braking system uses a forward-facing camera and forward-facing medium-to-long-range millimeter-wave radar to detect vehicles, pedestrians, and obstacles ahead, and automatically activates the vehicle's braking system to slow down the vehicle when a collision hazard may occur, so as to avoid a collision or mitigate the consequences of a collision.

[0080] In related technologies, due to the multipath effect of millimeter-wave radar and the susceptibility of cameras to weather and lighting conditions, false targets are easily generated simultaneously, leading to false fused targets. If the collision time of the false fused targets meets the AEB triggering condition, it will result in a false AEB trigger.

[0081] To overcome the above problems, this application proposes an emergency braking triggering method, which aims to remove false fusion targets by fusing target constraints, thereby avoiding false AEB triggering caused by false fusion targets.

[0082] refer to Figure 1 , Figure 1 This is a flowchart illustrating the steps of an emergency braking triggering method according to an embodiment of the present invention, as follows: Figure 1 As shown, the emergency braking triggering method includes:

[0083] Step S101: Obtain initial attribute information of multiple initial targets detected by millimeter-wave radar and camera.

[0084] In this embodiment, millimeter-wave radar and cameras are responsible for environmental perception. They can be used to detect the environment around the vehicle, such as vehicle information, road information, pedestrian information, etc. The detected information is then processed to obtain multiple initial targets and their corresponding initial attribute information, which is then transmitted to the vehicle controller. The vehicle controller can acquire the initial attribute information of the multiple initial targets detected by the millimeter-wave radar and cameras in real time. The initial targets can be vehicles or pedestrians, and the initial attribute information can be the lateral distance, longitudinal distance, lateral velocity, and longitudinal velocity of the initial targets. The number of initial targets detected by the millimeter-wave radar and cameras may be different.

[0085] Step S102: Based on the initial attribute information of the multiple initial targets, perform target fusion on the multiple initial targets detected by the millimeter-wave radar and the multiple initial targets detected by the camera to obtain multiple fused targets and fused attribute information corresponding to each fused target. The fused attribute information includes at least lateral distance, longitudinal distance, lateral velocity, longitudinal velocity and fused target survival time.

[0086] In this embodiment, target fusion is performed on multiple initial targets based on their initial attribute information, and initial targets with similar initial attribute information detected by millimeter-wave radar and camera are fused into a single fused target.

[0087] Specifically, based on the initial attribute information of the multiple initial targets, target fusion is performed on the multiple initial targets detected by the millimeter-wave radar and the multiple initial targets detected by the camera to obtain multiple fused targets and fused attribute information corresponding to each fused target, including the following steps:

[0088] The initial attribute information of multiple initial targets detected by the millimeter-wave radar and camera is parameterized to obtain the updated attribute information of multiple initial targets detected by the millimeter-wave radar and camera.

[0089] Based on the updated attribute information of the multiple initial targets, the multiple initial targets detected by the millimeter-wave radar are associated with the multiple initial targets detected by the camera to obtain the optimal association result between the multiple selected targets detected by the millimeter-wave radar and the multiple selected targets detected by the camera.

[0090] Based on the optimal association result, target fusion is performed on the multiple initial targets detected by the millimeter-wave radar and the multiple initial targets detected by the camera that are related to each other, to obtain multiple fused targets and fused attribute information corresponding to each fused target.

[0091] In this embodiment, since the millimeter-wave radar and camera are located at different positions and the detection data of the millimeter-wave radar and camera may be out of sync, in order to improve the accuracy of target fusion data, it is necessary to unify the initial attribute information of multiple initial targets detected by the millimeter-wave radar and camera. Specifically, this includes unifying the time and the coordinate system, so as to obtain the updated attribute information of multiple initial targets detected by the millimeter-wave radar and camera.

[0092] Based on the initial attribute information of multiple initial targets, the multiple initial targets detected by millimeter-wave radar and cameras are correlated to obtain the optimal correlation result between the multiple initial targets detected by millimeter-wave radar and the multiple initial targets detected by cameras. Specifically, for example, if the millimeter-wave radar detects five initial targets A1, A2, A3, A4, and A5, and the camera detects three targets B1, B2, and B3, then correlation means associating the same target (vehicle or pedestrian) detected by the millimeter-wave radar and camera. For example, A1 and B1 may be the same target, so A1 and B1 can be correlated. Correlating all possible initial targets with each other yields the optimal correlation result. Based on the optimal association result, target fusion is performed on multiple initial targets. That is, in the optimal association result, two mutually related initial targets are considered to be the same target, only detected by millimeter-wave radar and camera respectively. Therefore, it is necessary to fuse the same target detected by millimeter-wave radar and camera, and assign fused attribute information to the fused target. Specifically, the fused attribute information of the fused target includes the initial data and fused data corresponding to the fused target. The initial data is the initial attribute information detected by millimeter-wave radar and camera corresponding to the fused target. The fused lateral distance in the fused data can be the lateral distance detected by camera, the fused longitudinal distance in the fused data can be the longitudinal distance detected by millimeter-wave radar, the fused lateral velocity in the fused data can be the lateral velocity detected by camera, and the fused longitudinal velocity in the fused data can be the longitudinal velocity detected by millimeter-wave radar, thereby obtaining the fused attribute information corresponding to the fused target.

[0093] In one feasible implementation, the updated attribute information package includes the distance from the initial target to the vehicle, the standard deviation of the millimeter-wave radar velocity, and the number of stable tracking frames by the camera. Based on the updated attribute information of the multiple initial targets, the multiple initial targets detected by the millimeter-wave radar are associated with the multiple initial targets detected by the camera to obtain the optimal association result between the multiple selected targets detected by the millimeter-wave radar and the multiple selected targets detected by the camera, including:

[0094] The camera noise intensity value and the millimeter-wave radar noise intensity value are determined based on the distance from the initial target to the vehicle, the standard deviation of the millimeter-wave radar velocity, and the number of stable tracking frames of the camera.

[0095] Based on the camera noise intensity value and the millimeter-wave radar noise intensity value, the fusion weights of multiple initial targets detected by the millimeter-wave radar and the fusion weights of multiple initial targets detected by the camera are determined. The fusion weights of the multiple initial targets detected by the camera are negatively correlated with the camera noise intensity value, and the fusion weights of the multiple initial targets detected by the millimeter-wave radar are negatively correlated with the millimeter-wave radar noise intensity value.

[0096] Based on the fusion weights of the multiple initial targets detected by the millimeter-wave radar and the multiple initial targets detected by the camera, the optimal correlation result between the multiple initial targets detected by the millimeter-wave radar and the multiple initial targets detected by the camera is obtained.

[0097] In this embodiment, when calculating the fusion weight of the initial target using the fusion weight calculation strategy, it is necessary to calculate the fusion weight of multiple initial targets detected by millimeter-wave radar and the fusion weight of multiple initial targets detected by camera respectively. Specifically, it is necessary to calculate the weight of each attribute of each initial target detected by millimeter-wave radar and the weight of each attribute of each initial target detected by camera, thereby obtaining the fusion weight of multiple initial targets detected by millimeter-wave radar and the fusion weight of multiple initial targets detected by camera.

[0098] In this embodiment, the camera noise intensity value and the millimeter-wave radar noise intensity value are first determined based on the distance from the initial target to the vehicle, the standard deviation of the millimeter-wave radar velocity, and the number of stable tracking frames of the camera.

[0099] When millimeter-wave radar encounters external interference with stationary targets, such as fences or walls, its own resolution limitations may lead to larger measurement errors, increasing the standard deviation of millimeter-wave radar velocity measurement. This results in an increase in the lateral velocity error of the millimeter-wave radar. Moreover, the predicted value is slightly higher in this case, causing the error of the fused data to be slightly higher than the radar observation value, which may cause AEB to be falsely triggered.

[0100] To address the issue of false triggering caused by excessively large standard deviation in millimeter-wave radar speed measurement, a method can be adopted to appropriately reduce the measurement weight of the millimeter-wave radar while increasing the speed measurement weight of the camera.

[0101] Specifically, when the distance from the initial target to the vehicle is less than a preset distance, and the standard deviation of the millimeter-wave radar speed is greater than the preset standard deviation of the speed, and the number of stable tracking frames of the camera is greater than the preset number of tracking frames, the noise intensity value of the first camera and the noise intensity value of the first millimeter-wave radar are determined. The noise intensity value of the first camera is less than the default camera noise intensity value, and the noise intensity value of the first millimeter-wave radar is greater than the default camera noise intensity value.

[0102] The preset distance, preset speed standard deviation, and preset tracking frame count are all pre-input values. These values ​​ensure that the noise intensity value of the first camera is less than the default camera noise intensity value, and the noise intensity value of the first millimeter-wave radar is greater than the default camera noise intensity value. This increases the weight of camera measurements and decreases the weight of millimeter-wave radar measurements, thereby obtaining more accurate fusion weights for multiple selected targets detected by the millimeter-wave radar and multiple selected targets detected by the camera.

[0103] Then, based on the fusion weights of multiple initial targets detected by the millimeter-wave radar and the multiple initial targets detected by the camera, the KM algorithm is used to obtain the optimal association result between the multiple initial targets detected by the millimeter-wave radar and the multiple initial targets detected by the camera. The KM algorithm is a combinatorial optimization algorithm for solving the task assignment problem on a weighted graph. It is used to find a subgraph with disjoint vertices and an optimal weight sum. In the KM algorithm, the vertices are multiple initial targets, the edges are used to identify the association between the initial targets detected by the camera and the initial targets detected by the millimeter-wave radar, the fusion weight is Euclidean distance, and the optimal weight sum in the KM algorithm is the minimum weight sum when each selected target is associated at most once and the most selected associations are achieved.

[0104] In one feasible implementation, before the step of obtaining the optimal correlation result between the multiple initial targets detected by the millimeter-wave radar and the multiple initial targets detected by the camera based on the fusion weights of the multiple initial targets detected by the millimeter-wave radar and the fusion weights of the multiple initial targets detected by the camera, the method further includes:

[0105] For each of the multiple initial targets detected by the millimeter-wave radar, if the initial target is a single radar target, and the lateral distance of the initial target is less than a preset lateral distance, and the absolute value of the lateral velocity of the initial target is greater than a preset absolute value of the lateral velocity, and the initial target is in a recursive state, and the number of detection frames of the initial target is less than a preset number of frames, the initial target is eliminated.

[0106] In this embodiment, false triggering occurs when the millimeter-wave radar falsely detects a high-speed moving target. This type of false target occurs at close range, has a short tracking time, a large lateral speed, and its target detection state may be unstable. Therefore, multiple initial targets detected by the millimeter-wave radar can be eliminated according to a false target elimination strategy.

[0107] Among them, the preset lateral distance, the preset absolute value of lateral velocity, and the preset number of frames are all preset values. The recursive state is the current state of the initial target calculated based on the historical data of the initial target and the vehicle's own state, rather than the current state of the actual detected initial target. That is, at this time, no initial target is actually detected, the tracking time of the initial target is short, and the detection state of the initial target is unstable, so the initial target can be determined to be a false target and thus removed.

[0108] Step S103: Based on the fusion target constraints and the fusion attribute information corresponding to each fusion target, false targets are removed from the multiple fusion targets to obtain the true fusion targets.

[0109] In this embodiment, after obtaining multiple real fusion targets, the real fusion targets can be further screened according to the constraints of the fusion targets.

[0110] The fusion target constraint conditions include at least one of the following:

[0111] The difference between the lateral distances detected by the millimeter-wave radar and the camera is less than a preset lateral distance difference;

[0112] The survival time of the fusion target is greater than the first preset survival time;

[0113] The difference between the longitudinal distances detected by the millimeter-wave radar and the camera is less than a preset longitudinal distance difference;

[0114] The difference between the longitudinal velocity detected by the millimeter-wave radar and the camera is less than a preset longitudinal velocity difference;

[0115] When the fusion target is a vehicle, the survival time of the fusion target is greater than the second preset survival time; when the fusion target is a pedestrian, the survival time of the fusion target is greater than the third preset survival time, and the second preset survival time is greater than the third preset survival time.

[0116] All conditions apply to the same real fusion target. The lateral distance difference is the absolute value of the difference between the lateral distance detected by the millimeter-wave radar and the lateral distance detected by the camera. The survival time of the fusion target is the duration of continuous detection for the real fusion target. The longitudinal distance difference is the absolute value of the difference between the longitudinal distance detected by the millimeter-wave radar and the longitudinal distance detected by the camera. The longitudinal velocity difference is the absolute value of the difference between the longitudinal velocity detected by the millimeter-wave radar and the longitudinal velocity detected by the camera. Vehicles are more likely to be detected as false targets than pedestrians. By setting the second preset survival time to be greater than the third preset survival time, false vehicle targets can be better eliminated.

[0117] Step S104: Based on the fusion attribute information of each real fusion target, calculate the fusion collision time, the first initial collision time and the second initial collision time of each real fusion target.

[0118] In this embodiment, the fusion attribute information of each real fusion target includes fusion lateral distance, fusion longitudinal distance, fusion lateral velocity, and fusion longitudinal velocity. The fusion attribute information of each real fusion target also includes updated attribute information detected by the millimeter-wave radar and camera corresponding to that real fusion target. Based on the fusion lateral distance, fusion longitudinal distance, fusion lateral velocity, and fusion longitudinal velocity, the relative distance and relative velocity of the real fusion target can be calculated. The fusion collision time of each real fusion target can be obtained by directly dividing the relative distance by the relative velocity. The first initial collision time is the collision time calculated based on the updated attribute information detected by the millimeter-wave radar corresponding to the real fusion target. The second initial collision time... The collision time is calculated based on the updated attribute information detected by the camera corresponding to the real fused target. Specifically, the updated attribute information detected by the millimeter-wave radar and camera corresponding to the real fused target includes updated lateral distance, updated longitudinal distance, updated lateral velocity, and updated longitudinal velocity. The first initial collision time of the real fused target is calculated based on the updated lateral distance, updated longitudinal distance, updated lateral velocity, and updated longitudinal velocity detected by the millimeter-wave radar. The second initial collision time of the real fused target is calculated based on the updated lateral distance, updated longitudinal distance, updated lateral velocity, and updated longitudinal velocity detected by the camera. The specific calculation method can refer to the above calculation method for fusion collision time, and will not be repeated here.

[0119] Step S105: When the collision time of the real fusion target is less than the preset collision time, and the difference between the first initial collision time and the second initial collision time is less than the preset collision time difference, the emergency braking selection result is determined to be automatic emergency braking.

[0120] False fusion targets have been removed by the fusion target constraint. At this point, when the collision time of the real fusion target is less than the preset collision time, and the difference between the first initial collision time and the second initial collision time is less than the preset collision time difference, the emergency braking selection result can be determined as automatic emergency braking. The fact that the difference between the first initial collision time and the second initial collision time is less than the preset collision time difference can further remove false fusion targets, thereby avoiding false AEB triggering caused by false fusion targets meeting the AEB triggering conditions.

[0121] In the embodiments of this application, by increasing the weight of camera measurements and decreasing the weight of millimeter-wave radar measurements, a more accurate fusion weight of multiple initial targets detected by millimeter-wave radar and multiple initial targets detected by camera can be obtained. This avoids AEB false triggering caused by the fusion data error being slightly higher than the radar observation value. False targets (high-speed moving targets falsely detected by millimeter-wave radar) are also eliminated to avoid AEB false triggering. Finally, the fused targets are further screened and the constraints are tightened through fusion target constraints to avoid AEB false triggering.

[0122] Based on the same inventive concept, this application proposes an emergency braking triggering device, referring to... Figure 2 , Figure 2 This is a schematic diagram of an emergency braking triggering device according to an embodiment of the present invention, as shown below. Figure 2 As shown, the device includes:

[0123] The acquisition module 201 is used to acquire the initial attribute information of multiple initial targets detected by the millimeter-wave radar and camera;

[0124] The fusion module 202 is used to perform target fusion on multiple initial targets detected by the millimeter-wave radar and multiple initial targets detected by the camera based on the initial attribute information of the multiple initial targets, to obtain multiple fused targets and fused attribute information corresponding to each fused target. The fused attribute information includes at least lateral distance, longitudinal distance, lateral velocity, longitudinal velocity and fused target survival time.

[0125] The elimination module 203 is used to eliminate false targets from the multiple fusion targets according to the fusion target constraints and the fusion attribute information corresponding to each fusion target, so as to obtain the true fusion targets;

[0126] The calculation module 204 is used to calculate the fusion collision time, the first initial collision time and the second initial collision time of each real fusion target based on the fusion attribute information of each real fusion target.

[0127] The determination module 205 is used to determine the emergency braking selection result as automatic emergency braking when the collision time of the real fusion target is less than the preset collision time, and the difference between the first initial collision time and the second initial collision time is less than the preset collision time difference.

[0128] Optionally, the fusion target constraint includes at least one of the following:

[0129] The difference between the lateral distances detected by the millimeter-wave radar and the camera is less than a preset lateral distance difference;

[0130] The survival time of the fusion target is greater than the first preset survival time;

[0131] The difference between the longitudinal distances detected by the millimeter-wave radar and the camera is less than a preset longitudinal distance difference;

[0132] The difference between the longitudinal velocity detected by the millimeter-wave radar and the camera is less than a preset longitudinal velocity difference;

[0133] When the fusion target is a vehicle, the survival time of the fusion target is greater than the second preset survival time; when the fusion target is a pedestrian, the survival time of the fusion target is greater than the third preset survival time, and the second preset survival time is greater than the third preset survival time.

[0134] Optionally, the fusion module includes:

[0135] The unification submodule is used to unify the initial attribute information of multiple initial targets detected by the millimeter-wave radar and camera, and obtain the updated attribute information of multiple initial targets detected by the millimeter-wave radar and camera.

[0136] The association submodule is used to associate the multiple initial targets detected by the millimeter-wave radar with the multiple initial targets detected by the camera based on the updated attribute information of the multiple initial targets, so as to obtain the optimal association result between the multiple selected targets detected by the millimeter-wave radar and the multiple selected targets detected by the camera;

[0137] The fusion submodule is used to perform target fusion on the multiple initial targets detected by the millimeter-wave radar and the multiple initial targets detected by the camera that are related to each other, based on the optimal association result, so as to obtain multiple fused targets and fusion attribute information corresponding to each fused target.

[0138] Optionally, the computing module includes:

[0139] The acquisition submodule is used to acquire the updated attribute information detected by millimeter-wave radar and camera contained in the fusion attribute information of the real fusion target;

[0140] The first acquisition submodule is used to obtain the first initial collision time based on the updated attribute information detected by the millimeter-wave radar of the real fused target.

[0141] The second acquisition submodule is used to obtain the second initial collision time based on the updated attribute information detected by the camera of the real fusion target.

[0142] Optionally, the updated attribute information includes the distance from the initial target to the vehicle, the standard deviation of the millimeter-wave radar velocity, and the number of stable tracking frames by the camera;

[0143] The associated sub-modules include:

[0144] The first determining subunit is used to determine the camera noise intensity value and the millimeter-wave radar noise intensity value based on the distance from the initial target to the vehicle, the standard deviation of the millimeter-wave radar velocity, and the number of stable tracking frames of the camera.

[0145] The second determining subunit is used to determine the fusion weights of multiple initial targets detected by the millimeter-wave radar and the fusion weights of multiple initial targets detected by the camera based on the camera noise intensity value and the millimeter-wave radar noise intensity value. The fusion weights of the multiple initial targets detected by the camera are negatively correlated with the camera noise intensity value, and the fusion weights of the multiple initial targets detected by the millimeter-wave radar are negatively correlated with the millimeter-wave radar noise intensity value.

[0146] The association subunit is used to obtain the optimal association result between the multiple initial targets detected by the millimeter-wave radar and the multiple initial targets detected by the camera, based on the fusion weights of the multiple initial targets detected by the millimeter-wave radar and the fusion weights of the multiple initial targets detected by the camera.

[0147] Optionally, prior to the associated subunit, the device further includes:

[0148] The elimination subunit is used to eliminate each initial target among multiple initial targets detected by the millimeter-wave radar when the initial target is a single radar target, the lateral distance of the initial target is less than a preset lateral distance, the absolute value of the lateral velocity of the initial target is greater than a preset absolute value of the lateral velocity, the initial target is in a recursive state, and the number of detection frames of the initial target is less than a preset number of frames.

[0149] Optionally, the fusion module includes:

[0150] The first determining submodule is used to determine the lateral distance detected by the camera as the fusion lateral distance of the fusion target;

[0151] The first determining submodule is used to determine the longitudinal distance detected by the millimeter-wave radar as the fused longitudinal distance of the fused target;

[0152] The first determining submodule is used to determine the lateral velocity detected by the camera as the fusion lateral velocity of the fusion target;

[0153] The first determining submodule is used to determine the longitudinal velocity detected by the millimeter-wave radar as the fused longitudinal velocity of the fused target.

[0154] refer to Figure 3 , Figure 3 This is a schematic diagram of the structure of an electronic device according to an embodiment of the present invention, such as... Figure 3 As shown, this application also provides an electronic device, including:

[0155] Processor 31;

[0156] The device has a memory 32 storing instructions and a computer program stored on the memory 32 that can run on the processor 31. When the computer program is executed by the processor 31, it causes the device to perform an emergency braking triggering method.

[0157] This application also provides a non-transitory computer-readable storage medium storing a computer program that, when executed by a processor 31 of an electronic device, enables the electronic device to perform an emergency braking triggering method.

[0158] This application also provides a vehicle, including a vehicle body and an emergency braking triggering device disposed on the vehicle body, the emergency braking triggering device being used to perform the above-described emergency braking triggering method.

[0159] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0160] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, apparatus, or computer program products. Therefore, embodiments of the present invention can take the form of entirely hardware embodiments, entirely software embodiments, or embodiments combining software and hardware aspects. Furthermore, embodiments of the present invention can take the form of computer program products implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0161] Embodiments of the present invention are described with reference to flowchart illustrations and / or block diagrams of methods, terminal devices (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing terminal device to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing terminal device, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0162] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing terminal device to operate in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0163] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal equipment, causing a series of operational steps to be performed on the computer or other programmable terminal equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable terminal equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0164] Although preferred embodiments of the present invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of the embodiments of the present invention.

[0165] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or terminal device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or terminal device. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or terminal device that includes said element.

[0166] The above provides a detailed description of an emergency braking triggering method, device, electronic device, and vehicle provided by the present invention. Specific examples have been used to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. An emergency braking triggering method, characterized in that, Acquire initial attribute information of multiple initial targets detected by millimeter-wave radar and cameras; The initial attribute information of multiple initial targets detected by the millimeter-wave radar and camera is parameterized to obtain updated attribute information of multiple initial targets detected by the millimeter-wave radar and camera; the updated attribute information includes the distance from the initial target to the vehicle, the standard deviation of the millimeter-wave radar velocity, and the number of stable tracking frames of the camera; The camera noise intensity value and the millimeter-wave radar noise intensity value are determined based on the distance from the initial target to the vehicle, the standard deviation of the millimeter-wave radar velocity, and the number of stable tracking frames of the camera. Based on the camera noise intensity value and the millimeter-wave radar noise intensity value, the fusion weights of multiple initial targets detected by the millimeter-wave radar and the fusion weights of multiple initial targets detected by the camera are determined. The fusion weights of the multiple initial targets detected by the camera are negatively correlated with the camera noise intensity value, and the fusion weights of the multiple initial targets detected by the millimeter-wave radar are negatively correlated with the millimeter-wave radar noise intensity value. Based on the fusion weights of the multiple initial targets detected by the millimeter-wave radar and the multiple initial targets detected by the camera, the optimal correlation result between the multiple initial targets detected by the millimeter-wave radar and the multiple initial targets detected by the camera is obtained. Based on the optimal association result, target fusion is performed on the multiple initial targets detected by the millimeter-wave radar and the multiple initial targets detected by the camera that are related to each other, to obtain multiple fused targets and fused attribute information corresponding to each fused target. The fused attribute information includes at least lateral distance, longitudinal distance, lateral velocity, longitudinal velocity and fused target survival time. Based on the fusion target constraints and the fusion attribute information corresponding to each fusion target, false targets are eliminated from the multiple fusion targets to obtain the true fusion targets; Based on the fusion attribute information of each real fusion target, the fusion collision time, the first initial collision time and the second initial collision time of each real fusion target are calculated. When the collision time of the real fusion target is less than the preset collision time, and the difference between the first initial collision time and the second initial collision time is less than the preset collision time difference, the emergency braking selection result is determined to be automatic emergency braking. The first initial collision time and the second initial collision time of the real fusion target are obtained by the following method: The first initial collision time is obtained based on the updated attribute information detected by the millimeter-wave radar of the real fused target; The second initial collision time is obtained based on the updated attribute information detected by the camera of the real fusion target.

2. The emergency braking triggering method according to claim 1, characterized in that, The fusion objective constraint conditions include at least one of the following: The difference between the lateral distances detected by the millimeter-wave radar and the camera is less than a preset lateral distance difference; The survival time of the fusion target is greater than the first preset survival time; The difference between the longitudinal distances detected by the millimeter-wave radar and the camera is less than a preset longitudinal distance difference; The difference between the longitudinal velocity detected by the millimeter-wave radar and the camera is less than a preset longitudinal velocity difference; When the fusion target is a vehicle, the survival time of the fusion target is greater than the second preset survival time; When the fusion target is a pedestrian, the survival time of the fusion target is greater than the third preset survival time, and the second preset survival time is greater than the third preset survival time.

3. The emergency braking triggering method according to claim 1, characterized in that, Before the step of obtaining the optimal correlation result between the multiple initial targets detected by the millimeter-wave radar and the multiple initial targets detected by the camera based on the fusion weights of the multiple initial targets detected by the millimeter-wave radar and the fusion weights of the multiple initial targets detected by the camera, the method further includes: For each of the multiple initial targets detected by the millimeter-wave radar, if the initial target is a single radar target, and the lateral distance of the initial target is less than a preset lateral distance, and the absolute value of the lateral velocity of the initial target is greater than a preset absolute value of the lateral velocity, and the initial target is in a recursive state, and the number of detection frames of the initial target is less than a preset number of frames, the initial target is discarded.

4. The emergency braking triggering method according to claim 1, characterized in that, The fusion attribute information of the fusion target is obtained through the following method: The lateral distance detected by the camera is determined as the fusion lateral distance of the fusion target; The longitudinal distance detected by the millimeter-wave radar is determined as the fused longitudinal distance of the fused target; The lateral velocity detected by the camera is determined as the fusion lateral velocity of the fusion target; The longitudinal velocity detected by the millimeter-wave radar is determined as the fused longitudinal velocity of the fused target.

5. An emergency braking triggering device, characterized in that, include: The acquisition module is used to acquire initial attribute information of multiple initial targets detected by millimeter-wave radar and cameras; The fusion module is used to perform target fusion on multiple initial targets detected by the millimeter-wave radar and multiple initial targets detected by the camera based on the initial attribute information of the multiple initial targets, to obtain multiple fused targets and fused attribute information corresponding to each fused target. The fused attribute information includes at least lateral distance, longitudinal distance, lateral velocity, longitudinal velocity and fused target survival time. The elimination module is used to eliminate false targets from the multiple fusion targets based on the fusion target constraints and the fusion attribute information corresponding to each fusion target, so as to obtain the true fusion targets; The calculation module is used to calculate the fusion collision time, the first initial collision time and the second initial collision time of each real fusion target based on the fusion attribute information of each real fusion target. The determination module is used to determine the emergency braking selection result as automatic emergency braking when the collision time of the real fusion target is less than the preset collision time, and the difference between the first initial collision time and the second initial collision time is less than the preset collision time difference. The fusion module includes: The unification submodule is used to unify the initial attribute information of multiple initial targets detected by the millimeter-wave radar and camera, and obtain the updated attribute information of multiple initial targets detected by the millimeter-wave radar and camera. The association submodule is used to associate the multiple initial targets detected by the millimeter-wave radar with the multiple initial targets detected by the camera based on the updated attribute information of the multiple initial targets, so as to obtain the optimal association result between the multiple selected targets detected by the millimeter-wave radar and the multiple selected targets detected by the camera; The fusion submodule is used to perform target fusion on the multiple initial targets detected by the millimeter-wave radar and the multiple initial targets detected by the camera that are related to each other, based on the optimal association result, to obtain multiple fused targets and fusion attribute information corresponding to each fused target; The associated sub-modules include: The first determining subunit is used to determine the camera noise intensity value and the millimeter-wave radar noise intensity value based on the distance from the initial target to the vehicle, the standard deviation of the millimeter-wave radar velocity, and the number of stable tracking frames of the camera. The second determining subunit is used to determine the fusion weights of multiple initial targets detected by the millimeter-wave radar and the fusion weights of multiple initial targets detected by the camera based on the camera noise intensity value and the millimeter-wave radar noise intensity value. The fusion weights of the multiple initial targets detected by the camera are negatively correlated with the camera noise intensity value, and the fusion weights of the multiple initial targets detected by the millimeter-wave radar are negatively correlated with the millimeter-wave radar noise intensity value. The association subunit is used to obtain the optimal association result between the multiple initial targets detected by the millimeter-wave radar and the multiple initial targets detected by the camera, based on the fusion weights of the multiple initial targets detected by the millimeter-wave radar and the fusion weights of the multiple initial targets detected by the camera. The computing module includes: The acquisition submodule is used to acquire the updated attribute information detected by millimeter-wave radar and camera contained in the fusion attribute information of the real fusion target; The first acquisition submodule is used to obtain the first initial collision time based on the updated attribute information detected by the millimeter-wave radar of the real fused target. The second acquisition submodule is used to obtain the second initial collision time based on the updated attribute information detected by the camera of the real fusion target.

6. An electronic device, characterized in that, include: The memory, the processor, and the computer program stored in the memory and executable on the processor, wherein when executed by the processor, the computer program implements the steps of the emergency braking triggering method as described in any one of claims 1 to 4.

7. A vehicle, characterized in that, The system includes a vehicle body and an emergency braking triggering device disposed on the vehicle body, the emergency braking triggering device being used to perform the steps of the emergency braking triggering method as described in any one of claims 1 to 4.

Images