An automatic emergency braking method for vulnerable road users under complex conditions

Abstract

The present disclosure provides an automatic emergency braking method for vulnerable road users under complex working conditions, including: performing radar detection on the driving road of the target vehicle; if a target object is detected on the driving road of the target vehicle, predicting the first collision time of the target vehicle relative to the target object; if the first collision time is within a first preset time range, detecting the type of the target object based on a preset detection algorithm; if the type of the target object is a preset type, predicting the second collision time of the target vehicle relative to the target object; based on the relationship between the second collision time and the second preset time range, performing emergency braking on the target vehicle. Thus, considering the complex working conditions of driving, multiple time ranges are adopted to take into account both driving safety and vehicle driver characteristics, and emergency braking is performed on the driving vehicle under complex working conditions, effectively improving the driving safety of the vehicle.

Description

Technical Field

[0001] Embodiments of the present disclosure relate to the technical field of electric vehicles, and in particular, to an automatic emergency braking method applicable to vulnerable road users under complex working conditions. Background Art

[0002] In view of the special traffic conditions in China with heavy traffic pressure and complex road target composition, smart cars have increasingly stringent requirements on the performance and requirements of the Automatic Emergency Braking - Vulnerable Road Users (AEB-VRU) for vulnerable road users. Since the movements of road targets such as children, adult pedestrians, and cyclists are extremely random, it is difficult to model their trajectories. When faced with complex scenarios such as "ghost peeking" and "cross intersection turn avoidance", the system is prone to failure, resulting in collision avoidance failure, which brings safety hazards.

[0003] In addition, the research on AEB-VRU system is also an important part of the development of intelligent networked vehicles, which is of great significance to promoting the upgrading and transformation of the automobile industry, and can also promote the development of the automobile industry towards intelligence, networking, electrification and sharing. The traditional automatic emergency braking system can no longer adapt to the complex and changeable traffic scenes in China. It is urgent to develop an automatic emergency braking system that can not only meet the national conditions of my country's road traffic, but also protect vulnerable road users. Summary of the invention

[0004] The embodiments described herein provide an automatic emergency braking method for vulnerable road users under complex operating conditions, which overcomes the above-mentioned problems.

[0005] According to the present disclosure, an automatic emergency braking method for vulnerable road users under complex working conditions is provided, comprising:

[0006] Performing radar detection on the driving road of the target vehicle, wherein the radar detection is used to describe detecting whether there is a target object on the driving road of the target vehicle through radar signals;

[0007] If a target object is detected on the driving road of the target vehicle, a first collision time of the target vehicle relative to the target object is predicted, wherein the first collision time is used to describe the current collision time of the target vehicle relative to the target object when a target object is detected on the driving road of the target vehicle;

[0008] If the first collision duration is within a first preset duration range, detecting the type of the target object based on a preset detection algorithm;

[0009] If the type of the target object is a preset type, predicting a second collision time of the target vehicle relative to the target object, the preset type is used to describe the type corresponding to vulnerable road users, the second collision time is used to describe the predicted current collision time of the target vehicle relative to the target object when it is detected that the type of the target object is the preset type, and the second collision time is less than the first collision time;

[0010] Based on the relationship between the second collision time and a second preset time range, an emergency braking operation is performed on the target vehicle.

[0011] Optionally, performing an emergency braking operation on the target vehicle based on the relationship between the second collision duration and a second preset duration range includes:

[0012] If the second collision duration is within a second preset duration range, generating collision warning information;

[0013] Displaying the collision warning information so that the driver of the target vehicle can perform corresponding operations based on the collision warning information;

[0014] Among them, the maximum preset time length in the second preset time length range is smaller than the minimum preset time length in the first preset time length range.

[0015] Optionally, performing an emergency braking operation on the target vehicle based on the relationship between the second collision duration and a second preset duration range includes:

[0016] If the second collision duration is within a third preset duration, emergency braking is performed on the target vehicle;

[0017] Among them, the maximum preset time length in the third preset time length range is smaller than the minimum preset time length in the second preset time length range.

[0018] Optionally, performing an emergency braking operation on the target vehicle based on the relationship between the second collision duration and a second preset duration range includes:

[0019] If the second collision duration is not within the second preset duration range, determining that the driving level of the target vehicle is a safe driving level;

[0020] Continuing to predict a third collision time of the target vehicle relative to the target object, if it is determined that the third collision time is within the second preset time range, generating the collision warning information, and the third collision time is less than the second collision time;

[0021] or;

[0022] Continue to predict a third collision time of the target vehicle relative to the target object, and if it is determined that the third collision time is within the third preset time range, perform emergency braking on the target vehicle.

[0023] Optionally, before detecting the type of the target object based on a preset detection algorithm, the method further includes:

[0024] Get the preset network model;

[0025] Performing network structure processing on the preset network model, and optimizing the preset network model using driving information of vulnerable road users under complex working conditions, so as to determine the preset detection algorithm;

[0026] The network structure processing is used to describe adding a target detection layer to the original model structure of the preset network model.

[0027] Optionally, after detecting the type of the target object based on a preset detection algorithm, the method further includes:

[0028] If the type of the target object is not a preset type, the driving level of the target vehicle is determined to be a safe driving level.

[0029] Optionally, after predicting the first collision duration of the target vehicle relative to the target object, the method further includes:

[0030] If the first collision duration is not within the first preset duration range, it is determined that the driving level of the target vehicle is a safe driving level.

[0031] The automatic emergency braking method for vulnerable road users under complex working conditions provided in the embodiment of the present application performs radar detection on the driving road of the target vehicle, where the radar detection is used to describe whether there is a target object on the driving road of the target vehicle by detecting the radar signal; if the target object is detected on the driving road of the target vehicle, a first collision time of the target vehicle with respect to the target object is predicted, where the first collision time is used to describe the current collision time of the target vehicle with respect to the target object predicted when the target object is detected on the driving road of the target vehicle; if the first collision time is within a first preset time range, the type of the target object is detected based on a preset detection algorithm; if the type of the target object is a preset type, a second collision time of the target vehicle with respect to the target object is predicted, where the preset type is used to describe the type corresponding to the vulnerable road user, and the second collision time is used to describe the current collision time of the target vehicle with respect to the target object predicted when the type of the target object is detected to be a preset type, and the second collision time is less than the first collision time; based on the relationship between the second collision time and the second preset time range, an emergency braking operation is performed on the target vehicle. In this way, taking into account the complex driving conditions, multiple time ranges are adopted to take into account both driving safety and vehicle driver characteristics, and emergency braking is performed on the driving vehicle under complex conditions, effectively improving vehicle driving safety.

[0032] The above description is only an overview of the technical solution of the embodiment of the present application. In order to more clearly understand the technical means of the embodiment of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the embodiment of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the drawings of the embodiments will be briefly described below. It should be noted that the drawings described below only relate to some embodiments of the present disclosure, but are not intended to limit the present disclosure, wherein:

[0034] Figure 1 It is a flow chart of an automatic emergency braking method for vulnerable road users under complex working conditions provided by an embodiment of the present disclosure.

[0035] Figure 2 It is a structural schematic diagram of an automatic emergency braking system for vulnerable road users under complex working conditions provided by an embodiment of the present disclosure.

[0036] Figure 3 It is a flow chart of another automatic emergency braking method for vulnerable road users under complex working conditions provided by an embodiment of the present disclosure.

[0037] It should be noted that the elements in the drawings are schematic and not drawn to scale. DETAILED DESCRIPTION

[0038] In order to make the purpose, technical solution and advantages of the embodiments of the present disclosure clearer, the technical solution of the embodiments of the present disclosure will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present disclosure, rather than all of the embodiments. Based on the described embodiments of the present disclosure, all other embodiments obtained by those skilled in the art without creative work also fall within the scope of protection of the present disclosure.

[0039] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by a person skilled in the art to which the subject matter of the present disclosure belongs. It will be further understood that terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the specification and the relevant art, and will not be interpreted in an idealized or overly formal form unless otherwise explicitly defined herein. As used herein, a statement that two or more parts are "connected" or "coupled" together shall mean that the parts are joined together directly or through one or more intermediate components.

[0040] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase "embodiments" in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0041] The term "and / or" in this article is only a description of the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can mean: A exists, A and B exist at the same time, and B exists. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship. Terms such as "first" and "second" are only used to distinguish one component (or a part of a component) from another component (or another part of a component).

[0042] In the description of the present application, unless otherwise specified, “plurality” means more than two (including two), and similarly, “plurality groups” means more than two (including two).

[0043] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the accompanying drawings.

[0044] The automatic emergency braking method under complex working conditions for vulnerable road users provided in this embodiment is applied to the automatic emergency braking system under complex working conditions for vulnerable road users. The automatic emergency braking system under complex working conditions for vulnerable road users includes: VRU detection algorithm of environmental perception sensor, FMCW radar and visual sensor fusion model, AEB-VRU system modeling and complex scenario design, and AEB-VRU system verification under typical complex scenarios.

[0045] The AEB system for vulnerable road users spans two important branches of the intelligent vehicle field, environmental perception and decision control, and involves key technologies such as VRU target perception algorithm, sensor fusion technology, and collision avoidance control strategy. In view of the shortcomings of the current AEB-VRU system, such as poor real-time performance of sensor detection algorithms, single working scenarios, and control strategies that do not take into account driving safety and driver characteristics, from the perspective of intelligent driving, an automatic emergency braking system designed for vulnerable road users that integrates "people-vehicle-road-environment" was invented.

[0046] This embodiment establishes an automatic emergency braking system that integrates "people-vehicle-road-environment" and is designed for vulnerable road users. It enhances the real-time performance of the sensor detection algorithm, takes complex working conditions into consideration, and the control strategy takes into account both driving safety and driver characteristics, thereby effectively improving the driving safety of the vehicle.

[0047] Figure 1 is a flow chart of an automatic emergency braking method for vulnerable road users under complex working conditions provided by an embodiment of the present disclosure, such as Figure 1 As shown in FIG. 1 , the specific process of the automatic emergency braking method for vulnerable road users under complex conditions includes:

[0048] S110: Perform radar detection on the driving road of the target vehicle.

[0049] Among them, radar detection is used to describe the use of radar signals to detect whether there is a target object on the road where the target vehicle is traveling. By sending and receiving radar signals, it is possible to effectively detect whether there is a target object on the road where the target vehicle is traveling.

[0050] It should be noted that the target objects referred to in this embodiment may include: children, cyclists, elderly people, other moving vehicles, animals, etc.

[0051] S120: If a target object is detected on the driving road of the target vehicle, predict a first collision duration of the target vehicle with respect to the target object.

[0052] The first collision duration is used to describe the predicted current collision duration of the target vehicle relative to the target object when a target object is detected on the driving road of the target vehicle.

[0053] In some embodiments, after predicting the first collision duration of the target vehicle with respect to the target object, the method further includes:

[0054] If the first collision duration is not within the first preset duration range, it is determined that the driving level of the target vehicle is a safe driving level.

[0055] Among them, when the first collision duration is not within the first preset duration range, it is considered that there is no driving danger in the range where the target object is located, and the driving level of the target vehicle is determined to be a safe driving level, that is, it will not cause danger to the driving of the target object.

[0056] S130: If the first collision duration is within a first preset duration range, detecting the type of the target object based on a preset detection algorithm.

[0057] Among them, the preset detection algorithm is a target detection algorithm based on the camera of the target vehicle, which is used to identify whether the type of the target object is the type corresponding to the ARU.

[0058] In some embodiments, before detecting the type of the target object based on a preset detection algorithm, the method further includes:

[0059] Obtain a preset network model; perform network structure processing on the preset network model, and use driving information of vulnerable road users under complex working conditions to optimize the preset network model to determine a preset detection algorithm.

[0060] Among them, network structure processing is used to describe adding a target detection layer to the original model structure of the preset network model.

[0061] The premise of achieving good performance of the AEB-VRU system is to obtain accurate environmental information. By reasonably selecting the system perception sensor, in view of the defects of the original YOLOv5s model in generating false detection and missed detection of small targets and insufficient VRU feature extraction capability, this embodiment adds a small target detection layer on the basis of the network layer of the original YOLOv5s model structure (i.e., the preset network model) to improve the YOLOv5 algorithm and thus improve the feature layer fusion effect. At the same time, the CIoU_Loss and DIoU_NMS methods are introduced to optimize the prediction end, thereby improving the detection performance of small targets and overlapping targets such as VRU.

[0062] In this embodiment, the preset types of targets are mainly vulnerable road groups such as pedestrians, cyclists, and motorcyclists, while passenger cars, trucks, buses, and other targets on the road are the main groups that pose a security threat to VRUs. Taking into account the huge number of cars and the complex road traffic flow, appropriate pictures are selected and labeled from public data sets such as Caltech pedestrian database, Daimler database, Tsinghua-Daimler dataset, Cityscapes dataset, Penn-Fudan database, and nuScenes as training sets and test sets for the improved YOLOv5 algorithm. The model is trained by transfer learning. The mAP training accuracy of the improved YOLOv5 algorithm reaches 90.2%, which is 4.5% higher than that of the original algorithm, further improving the detection of small target objects such as pedestrians and cyclists, and has the advantages of high detection accuracy and extremely fast detection speed.

[0063] By establishing and verifying the fusion model of radar and camera, the experiment selected various driving scenes such as suburban areas, urban roads, inner ring highways, tunnels, etc. to test the preset detection algorithm. The improved YOLOv5 algorithm has excellent detection effects on vehicles, pedestrians, cyclists and other targets in the above scenes, and does not misidentify trees, flowers, buildings and other targets that appear in the scene. Even in the complex intersection of people and vehicles in urban blocks and dark tunnel scenes, the algorithm has shown strong robustness and generalization ability.

[0064] In some embodiments, after detecting the type of the target object based on a preset detection algorithm, the method further includes:

[0065] If the type of the target object is not a preset type, the driving level of the target vehicle is determined to be a safe driving level.

[0066] Among them, if the type of the target object is not a preset type, it is considered that there is no driving danger within the range of the target vehicle, and the driving level of the target vehicle is determined to be a safe driving level, that is, it will not cause danger to the normal driving of the target vehicle.

[0067] S140: If the type of the target object is a preset type, predict a second collision duration of the target vehicle relative to the target object.

[0068] Among them, the preset type is used to describe the type corresponding to the vulnerable road user, and the second collision time is used to describe the predicted current collision time of the target vehicle relative to the target object when the type of the target object is detected to be the preset type. The second collision time is less than the first collision time.

[0069] S150: Based on the relationship between the second collision duration and the second preset duration range, perform emergency braking on the target vehicle.

[0070] Among them, the boundary durations of the second preset duration range and the first preset duration range may be different, that is, the maximum duration of the second preset duration range and the minimum duration of the first preset duration range are not continuous values, such as the first preset duration range is: 10s-15s, and the second preset duration range is: 5s-8s.

[0071] In some embodiments, based on the relationship between the second collision duration and the second preset duration range, performing an emergency braking operation on the target vehicle includes:

[0072] If the second collision duration is within the second preset duration range, a collision warning message is generated; and the collision warning message is displayed so that the driver of the target vehicle can perform corresponding operations based on the collision warning message.

[0073] The maximum preset duration in the second preset duration range is smaller than the minimum preset duration in the first preset duration range.

[0074] Therefore, when the second collision duration is within the second preset duration range, a warning is given to the driver of the vehicle to alert the driver to take actions such as slowing down or changing lanes.

[0075] In some other embodiments, based on the relationship between the second collision duration and the second preset duration range, emergency braking is performed on the target vehicle, including:

[0076] If the second collision duration is within the third preset duration range, emergency braking is performed on the target vehicle, and the maximum preset duration in the third preset duration range is less than the minimum preset duration in the second preset duration range, so as to avoid or mitigate collision accidents to the greatest extent.

[0077] The boundary durations of the third preset duration range and the second preset duration range may be different, that is, the maximum duration of the third preset duration range and the minimum duration of the second preset duration range are not continuous values, such as the second preset duration range is: 5s-8s, and the third preset duration range is: 1s-3s.

[0078] In some other embodiments, based on the relationship between the second collision duration and the second preset duration range, emergency braking is performed on the target vehicle, including:

[0079] If the second collision time is not within the second preset time range, the driving level of the target vehicle is determined to be a safe driving level; continue to predict the third collision time of the target vehicle relative to the target object, if it is determined that the third collision time is within the second preset time range, generate collision warning information, and the third collision time is less than the second collision time; or; continue to predict the third collision time of the target vehicle relative to the target object, if it is determined that the third collision time is within the third preset time range, perform emergency braking on the target vehicle.

[0080] Therefore, by continuously detecting the current collision duration of the target vehicle relative to the target object at different driving moments, it is convenient to take corresponding measures in time to avoid driving safety problems.

[0081] In this embodiment, radar detection is performed on the target vehicle's driving road. Radar detection is used to describe whether there is a target object on the target vehicle's driving road through radar signals; if a target object is detected on the target vehicle's driving road, the first collision time of the target vehicle relative to the target object is predicted. The first collision time is used to describe the current collision time of the target vehicle relative to the target object when the target object is detected on the target vehicle's driving road; if the first collision time is within the first preset time range, the type of the target object is detected based on the preset detection algorithm; if the type of the target object is the preset type, the second collision time of the target vehicle relative to the target object is predicted. The preset type is used to describe the type corresponding to the vulnerable road user. The second collision time is used to describe the current collision time of the target vehicle relative to the target object when the type of the target object is detected as the preset type. The second collision time is less than the first collision time; based on the relationship between the second collision time and the second preset time range, the target vehicle is subjected to emergency braking. In this way, considering the complex working conditions of driving, multiple time ranges are adopted to take into account both driving safety and vehicle driver characteristics, and emergency braking is performed on the driving vehicle under complex working conditions, effectively improving vehicle driving safety.

[0082] like Figure 2 As shown, the automatic emergency braking system for vulnerable road users under complex working conditions provided by this embodiment includes: a VRU detection algorithm of an environmental perception sensor, a FMCW radar and visual sensor fusion model, AEB-VRU system modeling and complex scenario design, and AEB-VRU system verification in typical complex scenarios.

[0083] In terms of environmental perception, the following Figure 3The decision-level fusion strategy of the shown sensor. Based on the predicted time to collision (TTC) of the radar signal with the target object, if the TTC is less than a certain threshold, the object is considered dangerous. At this time, the camera-based target detection algorithm starts to run to determine whether the detected target is a vulnerable road user (VRU) target such as a pedestrian; if it is and the TTC drops to another threshold, the AEB-VRU system activates the driver collision warning function to alert the driver so that they can make actions such as decelerating or changing lanes; finally, if the TTC drops to the last threshold, the system implements active braking to avoid or mitigate the collision accident to the greatest extent.

[0084] For example, the decision-level fusion strategy of the sensor can be implemented in the following steps.

[0085] S301. The radar detects the target object.

[0086] S302. Calculate the time to collision TTC (expressed as a duration).

[0087] S303. Determine whether TTC0 < TTC < TTC1. If so, execute S304; if not, execute S305.

[0088] S304. Run the camera target detection algorithm.

[0089] S305. The vehicle is at a safe driving level.

[0090] S306. Determine whether the target object type is a VRU target. If so, execute S307; if not, execute S305.

[0091] S307. Calculate the time to collision of the VRU target.

[0092] S308. Determine whether TTC2 < TTC < TTC3. If so, execute S309; if not, execute S305.

[0093] S309. Determine whether TTC4 < TTC < TTC5. If so, execute S310; if not, execute S311.

[0094] S310. The active system works.

[0095] S311. The collision warning system works.

[0096] When modeling the AEB-VRU system and designing complex scenarios, we took the pedestrian AEB system as the basis, expanded it to children and cyclists as the target, carried out trajectory planning and tracking control for complex scenarios, and carried out overall functional design of the AEB-VRU system, including system dynamics modeling, collision warning system design, safety assessment model establishment, collision avoidance control strategy establishment, etc. A vehicle dynamics model was established based on PreScan, and an inverse dynamics acceleration / braking logic switching model was built in MATLAB / Simulink; an evaluation model integrating safety distance and remaining collision time was proposed, hierarchical control was realized by using feedforward neural network and lower-level PID algorithm, a new fitness function was designed to evaluate the control system response, and the simulated annealing algorithm was used to improve the parameter setting efficiency and control accuracy of the control system; sampling compensation of sensor data and a position prediction method based on Kalman filtering were proposed for the "ghost poking" scenario, and path planning was carried out under the condition of the intersection, and trajectory tracking was carried out based on the optimal multi-point preview control.

[0097] Verify the AEB-VRU system in typical complex scenarios. PreScan is responsible for providing the test scenario and vehicle dynamics module, encapsulating scene elements such as roads, vehicles, pedestrians, cyclists, sensors, etc. into MATLAB / Simulink and building corresponding control models, including millimeter-wave radar and camera fusion models, collision warning subsystems, hierarchical control modules, etc., to verify and analyze the system.

[0098] Taking the VRU test protocol in the 2021 version of Euro NCAP as the benchmark, three complex driving scenarios of CBNAO, CPNCO and CPTA were determined for simulation testing of the system. A joint simulation platform was built based on PreScan and MATLAB / Simulink, and different driving speeds were selected to conduct in-depth analysis of the test results on different road surfaces. The results show that the system can adapt to the dynamic changes of different road surfaces, and actively avoid collisions with road targets on high and medium adhesion coefficient roads. The final redundant safety distance is stable at about 0.6-2.2m, and the collision speed can be minimized on low adhesion roads, reducing the risk of serious injury and death of VRU; all subsystems of the system can work in a coordinated manner without interfering with the driver's normal driving, the collision warning system has no false alarms or missed alarms, the hierarchical control system tracks the expected acceleration well, and the steady-state error is small, indicating that the designed AEB-VRU system integrating "human-vehicle-road-environment" has the advantages of strong adaptability to working conditions, good robustness, stable and reliable operation.

[0099] Each functional unit or module in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0100] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) or a processor (processor) to execute all or part of the steps of the various embodiments of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, etc., and other media that can store program codes.

[0101] In the claims, any reference symbols placed between brackets shall not be construed as limiting the claims. The word "comprising" described in the present application does not exclude the presence of elements or steps not listed in the claims. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The present application can be implemented with the aid of hardware comprising several different elements and with the aid of a suitably programmed computer. In a unit claim that lists a number of devices, several units of these devices may be embodied by the same hardware item. The use of first, second, and third, etc. does not indicate any order, and these words may be interpreted as names. The steps in the above embodiments, unless otherwise specified, should not be understood as limitations on the order of execution.

[0102] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. An automatic emergency braking method for vulnerable road users under complex working conditions, characterized in that: include: Performing radar detection on the driving road of the target vehicle, wherein the radar detection is used to describe detecting whether there is a target object on the driving road of the target vehicle through radar signals; If a target object is detected on the driving road of the target vehicle, a first collision time of the target vehicle relative to the target object is predicted, wherein the first collision time is used to describe the current collision time of the target vehicle relative to the target object when a target object is detected on the driving road of the target vehicle; If the first collision duration is within a first preset duration range, detecting the type of the target object based on a preset detection algorithm; If the type of the target object is a preset type, predicting a second collision time of the target vehicle relative to the target object, the preset type is used to describe the type corresponding to vulnerable road users, the second collision time is used to describe the predicted current collision time of the target vehicle relative to the target object when it is detected that the type of the target object is the preset type, and the second collision time is less than the first collision time; Based on the relationship between the second collision duration and a second preset duration range, performing an emergency braking operation on the target vehicle; If the type of the target object is not a preset type, the driving level of the target vehicle is determined to be a safe driving level.

2. The method according to claim 1, characterized in that The performing an emergency braking operation on the target vehicle based on the relationship between the second collision duration and the second preset duration range includes: If the second collision duration is within a second preset duration range, generating collision warning information; Displaying the collision warning information so that the driver of the target vehicle can perform corresponding operations based on the collision warning information; Among them, the maximum preset time length in the second preset time length range is smaller than the minimum preset time length in the first preset time length range.

3. The method according to claim 2, characterized in that The performing an emergency braking operation on the target vehicle based on the relationship between the second collision duration and the second preset duration range includes: If the second collision duration is within a third preset duration, emergency braking is performed on the target vehicle; Among them, the maximum preset time length in the third preset time length range is smaller than the minimum preset time length in the second preset time length range.

4. The method according to claim 3, characterized in that The performing an emergency braking operation on the target vehicle based on the relationship between the second collision duration and the second preset duration range includes: If the second collision duration is not within the second preset duration range, determining that the driving level of the target vehicle is a safe driving level; Continuing to predict a third collision time of the target vehicle relative to the target object, if it is determined that the third collision time is within the second preset time range, generating the collision warning information, and the third collision time is less than the second collision time; or; Continue to predict a third collision time of the target vehicle relative to the target object, and if it is determined that the third collision time is within the third preset time range, perform emergency braking on the target vehicle.

5. The method according to claim 1, characterized in that Before detecting the type of the target object based on a preset detection algorithm, the method further includes: Get the preset network model; Performing network structure processing on the preset network model, and optimizing the preset network model using driving information of vulnerable road users under complex working conditions, so as to determine the preset detection algorithm; The network structure processing is used to describe adding a target detection layer to the original model structure of the preset network model.

6. The method according to claim 1, characterized in that After predicting the first collision duration of the target vehicle relative to the target object, the method further includes: If the first collision duration is not within the first preset duration range, it is determined that the driving level of the target vehicle is a safe driving level.

Images