Method for screening vulnerable traffic participants

Abstract

The invention provides a method for screening vulnerable traffic participants, and belongs to the technical field of automatic driving environment perception and decision planning, and the method comprises the steps: obtaining the vulnerable traffic participants after preliminary screening through the state information of the vulnerable traffic participants; the state information comprises a historical track and a current position; the current position comprises a transverse distance and a longitudinal distance between the vulnerable traffic participant and the vehicle; obtaining vulnerable traffic participants after secondary screening by utilizing the rationality of the historical track and the overlapping degree of the historical track and the lane range; and quantifying the threat degree of the vulnerable traffic participants subjected to secondary screening by using the state information of the vulnerable traffic participants so as to take the vulnerable traffic participants with the threat degree greater than a preset threat degree threshold as actual vulnerable traffic participants. According to the method, the trajectory prediction result of the vulnerable traffic participant is evaluated, and target screening is completed, so that target misjudgment and unnecessary vehicle control intervention are reduced.

Description

Technical Field

[0001] This invention relates to the field of autonomous driving environmental perception and decision-making planning technology, and in particular to a method for screening vulnerable traffic participants. Background Technology

[0002] The application of autonomous driving is gradually expanding from highways to urban areas. In urban autonomous driving scenarios, the behavior of vulnerable road users is highly random and unpredictable. Existing technologies have the following problems: the number of VRUs detected by sensors is huge, but most of them have no substantial impact on the vehicle, resulting in a waste of computing resources; traditional screening methods based on distance or speed are difficult to distinguish between potential threat targets (such as pedestrians crossing the road) and harmless targets (such as stationary pedestrians on the roadside). Summary of the Invention

[0003] The present invention aims to solve at least one of the technical problems existing in the prior art, and proposes a method for screening vulnerable traffic participants in a preprocessing set of vulnerable traffic participants.

[0004] In a first aspect, embodiments of the present invention provide a method for screening vulnerable traffic participants, comprising: obtaining vulnerable traffic participants after initial screening using the status information of vulnerable traffic participants; the status information includes historical trajectory and current location; the current location includes the lateral distance and longitudinal distance between the vulnerable traffic participant and the vehicle; obtaining vulnerable traffic participants after secondary screening using the rationality of the historical trajectory and the overlap between the historical trajectory and the lane range; and quantifying the threat level of the vulnerable traffic participants after secondary screening using the status information of the vulnerable traffic participants so as to identify vulnerable traffic participants whose threat level is greater than a preset threat level threshold as actual vulnerable traffic participants.

[0005] In an embodiment of the present invention, the vulnerable traffic participants are obtained by utilizing their state information, including: extracting the time point and direction of change of movement based on the historical trajectory of the vulnerable traffic participants; filtering the vulnerable traffic participants if they meet a first distance condition; the distance condition is that the longitudinal distance is less than a preset minimum longitudinal distance threshold, or the lateral distance is greater than a preset maximum lateral distance threshold, or the longitudinal distance is greater than a first preset threshold and the time of the longitudinal distance being greater than the first preset threshold is greater than a second preset threshold; retaining the vulnerable traffic participants if they meet a second distance condition among the filtered vulnerable traffic participants; the second distance condition is that the lateral distance is less than a third preset threshold, or the difference between the current movement direction and the historical average movement direction is less than a fourth preset threshold; wherein, longitudinal is the direction along the tangent of the lane centerline; lateral is the direction perpendicular to the lane centerline.

[0006] In an embodiment of the present invention, the historical trajectory of a vulnerable traffic participant is used to extract the time point and direction of movement where the direction of movement changes. Prior to this, the method further includes: adjusting preset configuration parameters, specifically including: the preset configuration parameters include a preset maximum lateral distance threshold and a preset minimum longitudinal distance threshold; determining whether the current location is an intersection; if so, increasing the preset maximum lateral distance threshold and decreasing the preset minimum longitudinal distance threshold; if not, using the preset maximum lateral distance threshold and the preset minimum longitudinal distance threshold.

[0007] In an embodiment of the present invention, the reasonableness of historical trajectories and the degree of overlap between historical trajectories and lane ranges are used to obtain vulnerable traffic participants after two screenings, including: judging the reasonableness of historical trajectories based on lateral distance and lateral error between every two adjacent historical trajectory points, and retaining vulnerable traffic participants with reasonable historical trajectories; in response to the overlap between historical trajectories and lane ranges or the current location being an intersection, vulnerable traffic participants are retained.

[0008] In an embodiment of the present invention, the rationality of a historical trajectory is determined based on the lateral distance and the lateral error between every two adjacent historical trajectory points, including: retaining a vulnerable traffic participant if the lateral distance is less than a fifth preset threshold; calculating a weighted sum of the lateral errors between every two adjacent historical trajectory points; obtaining an average lateral error based on the weighted sum and the number of historical trajectory points; and retaining a vulnerable traffic participant if the average lateral error is less than a sixth preset threshold.

[0009] In an embodiment of the present invention, the threat level of vulnerable traffic participants after two screenings is quantified using the status information of vulnerable traffic participants. This includes: cropping the historical trajectory of each vulnerable traffic participant after two screenings to obtain a set of trajectory points within a preset time period; and calculating the threat level using a weighted sum of the reciprocal of the longitudinal distance of the vulnerable traffic participant, the ratio of lateral speed to lateral distance, and the information entropy of the trajectory points of the historical trajectory.

[0010] In an embodiment of the present invention, the vulnerable traffic participants after a second screening are obtained by utilizing the rationality of historical trajectories and the degree of overlap between historical trajectories and lane ranges. Prior to this, the method further includes: determining whether the vulnerable traffic participants after the initial screening are located on zebra crossings; if so, the vulnerable traffic participants are taken as actual vulnerable traffic participants; if not, the method of obtaining the vulnerable traffic participants after a second screening by utilizing the rationality of historical trajectories and the degree of overlap between historical trajectories and lane ranges is performed on the vulnerable traffic participants.

[0011] The second aspect of this disclosure provides a screening system for vulnerable traffic participants, which can be used to implement the aforementioned screening method for vulnerable traffic participants, including: a preliminary screening module, used to obtain vulnerable traffic participants after preliminary screening using the status information of the vulnerable traffic participants; the status information includes historical trajectory and current location; the current location includes the lateral distance and longitudinal distance between the vulnerable traffic participant and the vehicle; a secondary screening module, used to obtain vulnerable traffic participants after secondary screening using the rationality of the historical trajectory and the overlap between the historical trajectory and the lane range; and a quantification module, used to quantify the threat level of the vulnerable traffic participants after secondary screening using the status information of the vulnerable traffic participants so as to identify vulnerable traffic participants whose threat level is greater than a preset threat level threshold as actual vulnerable traffic participants.

[0012] A third aspect of this disclosure provides an electronic device comprising: one or more processors; and a memory for storing one or more programs, wherein, when the one or more programs are executed by the one or more processors, the one or more processors perform the aforementioned method for screening vulnerable traffic participants.

[0013] A fourth aspect of this disclosure also provides a computer-readable storage medium having executable instructions stored thereon, which, when executed by a processor, cause the processor to perform the aforementioned method for screening vulnerable traffic participants.

[0014] The method for screening vulnerable traffic participants provided by this invention filters out non-critical targets through a multi-level joint screening mechanism and trajectory credibility assessment, thereby improving computational efficiency and at least partially solving the technical problem of pedestrian / non-motorized vehicle interaction decision-making in complex urban scenarios. Attached Figure Description

[0015] Figure 1 A flowchart illustrating the screening method for vulnerable traffic participants provided in an embodiment of the present invention;

[0016] Figure 2 This is a schematic diagram of a process including zebra crossing screening provided in an embodiment of the present invention;

[0017] Figure 3 This is a flowchart illustrating an optional specific implementation method of step S2 in an embodiment of the present invention;

[0018] Figure 4 A structural block diagram of a screening system for vulnerable traffic participants provided in an embodiment of the present invention;

[0019] Figure 5 This is a structural block diagram of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0020] To enable those skilled in the art to better understand the technical solutions of the present invention, exemplary embodiments of the present invention are described below in conjunction with the accompanying drawings, including various details of the embodiments of the present invention to aid understanding. These should be considered merely exemplary. Therefore, those skilled in the art should recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the present invention. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0021] Where there is no conflict, the various embodiments of the present invention and the features thereof may be combined with each other.

[0022] As used herein, the term “and / or” includes any and all combinations of one or more related enumerated entries.

[0023] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used herein, the singular forms “a” and “the” are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will also be understood that when the terms “comprising” and / or “made of” are used in this specification, the presence of the stated feature, integral, step, operation, element, and / or component is specified, but the presence or addition of one or more other features, integrals, steps, operations, elements, components, and / or groups thereof is not excluded. Terms such as “connected” or “linked” are not limited to physical or mechanical connections but can include electrical connections, whether direct or indirect.

[0024] Unless otherwise specified, all terms used herein (including technical and scientific terms) have the same meaning as commonly understood by one of ordinary skill in the art. It will also be understood that terms such as those defined in commonly used dictionaries should be interpreted as having the meaning consistent with their meaning in the context of the relevant art and the invention, and will not be interpreted as having an idealized or overly formal meaning unless expressly so defined herein.

[0025] In the technical solution of this invention, the collection, storage, use, processing, transmission, provision, and disclosure of user personal information all comply with relevant laws and regulations and do not violate public order and good morals. The use of user data in this technical solution follows relevant national laws and regulations (e.g., the "Information Security Technology - Personal Information Security Specification"). For example: appropriate measures are taken for personal information access control; restrictions are imposed on the display of personal information; the purpose of using personal information does not exceed the scope of direct or reasonable association; and explicit identity targeting is eliminated when using personal information to avoid precisely locating a specific individual.

[0026] The application of autonomous driving is gradually expanding from highways to urban areas. Urban intersections are complex, and compared to highway scenarios, they include vulnerable road users such as pedestrians and cyclists. Autonomous driving systems need to fully consider the behavior prediction of these vulnerable road users and further decide whether to react accordingly. This invention proposes a method for evaluating the trajectory prediction results of vulnerable road users and completing target selection, thereby reducing target misjudgment and unnecessary vehicle control intervention.

[0027] Figure 1 A flowchart illustrating the screening method for vulnerable traffic participants provided in this embodiment of the invention; as shown. Figure 1 As shown, this embodiment of the invention provides a method for screening vulnerable traffic participants, comprising: S1, using the status information of vulnerable traffic participants to obtain initially screened vulnerable traffic participants; the status information includes historical trajectory and current location; the current location includes the lateral distance and longitudinal distance between the vulnerable traffic participant and the vehicle; S2, using the rationality of the historical trajectory and the overlap between the historical trajectory and the lane range to obtain second-screened vulnerable traffic participants; S3, using the status information of vulnerable traffic participants to quantify the threat level of the second-screened vulnerable traffic participants so that vulnerable traffic participants with a threat level greater than a preset threat level threshold are identified as actual vulnerable traffic participants.

[0028] This invention proposes a method for evaluating trajectory detection results of vulnerable road users (VRUs) and performing target screening. The initial screening quickly reduces the number of VRUs, while a second screening further filters based on trajectory rationality and lane correlation, reducing the probability of false positives. Furthermore, hazard quantification identifies high-threat targets, improving the safety of path planning. This method aims to determine whether detected VRUs within the detection range should be included in the autonomous driving system's considerations to assess their potential impact on path planning and decision-making, thereby improving the efficiency and safety of path planning.

[0029] Based on the above embodiments, the vulnerable traffic participants are obtained by utilizing their status information, including: extracting the time points and directions of change of movement direction using the historical trajectories of the vulnerable traffic participants; filtering out vulnerable traffic participants that meet the first distance condition; the distance condition is that the longitudinal distance is less than a preset minimum longitudinal distance threshold, or the lateral distance is greater than a preset maximum lateral distance threshold, or the longitudinal distance is greater than a first preset threshold and the time for the longitudinal distance to be greater than the first preset threshold is greater than a second preset threshold; retaining vulnerable traffic participants that meet the second distance condition among the filtered vulnerable traffic participants; the second distance condition is that the lateral distance is less than a third preset threshold, or the difference between the current movement direction and the historical average movement direction is less than a fourth preset threshold; wherein, longitudinal is the direction along the tangent of the lane centerline; lateral is the direction perpendicular to the lane centerline.

[0030] In this embodiment, firstly, the location, direction of movement, and historical trajectory of the VRUs are analyzed to identify those requiring further processing. The filtering logic combines multiple conditions (such as distance, time interval, and directional consistency) to ensure the accuracy of the filtering results.

[0031] The specific steps are as follows:

[0032] (1) Obtain configuration parameters and historical trajectory data. Obtain the historical trajectory data of the target and determine whether the current state is an intersection stage. If the current state is an intersection, increase the maximum lateral distance threshold and decrease the minimum longitudinal distance threshold to broaden the scope of consideration.

[0033] (2) Directional data storage and retrieval. The historical trajectory directional data of the VRU is stored in a double-ended queue. The historical trajectory data is traversed to extract the motion direction (θ) at each time point.

[0034] (3) Conditional judgment. If the longitudinal distance Less than the minimum threshold or horizontal distance Greater than the maximum threshold or longitudinal distance Greater than the threshold and time interval Greater than the threshold If the longitudinal distance is less than the minimum threshold, and the VRU is already very close to the side or rear of the vehicle, it is already being handled by other emergency systems, and there is no need to track it as a long-term threat, avoiding duplicate processing or conflicting decisions. If the lateral distance is greater than the maximum threshold, and the VRU is located in an adjacent lane or on the roadside, with no possibility of intersecting with the vehicle's expected path, then this safety target is excluded. If the longitudinal distance... Greater than the threshold and time interval Greater than the threshold Therefore, it would take a very long time for the VRU and the vehicle to meet, so this option is also excluded.

[0035] If horizontal distance or current direction of movement relative to historical average direction The difference is less than the threshold If the VRU is close to or has entered its own lane, then retain the VRU. If the VRU is already close to or has entered its own lane, there is a high risk of collision, and it needs to be retained. In addition, even if the VRU is not currently close in lateral distance, but its direction of movement is very continuous and changes significantly, the target also needs to be retained.

[0036] Through the embodiments of the present invention, obviously harmless VRUs (such as stationary pedestrians on the roadside) can be quickly removed, reducing the waste of computing resources, while ensuring that potential threat targets (such as pedestrians crossing the road) are not mistakenly filtered.

[0037] Based on the above embodiments, the historical trajectory of vulnerable traffic participants is used to extract the time point and direction of movement where the direction of movement changes. Before this, the method also includes: adjusting preset configuration parameters, specifically including: preset configuration parameters include preset maximum lateral distance threshold and preset minimum longitudinal distance threshold; determining whether the current location is an intersection; if so, increasing the preset maximum lateral distance threshold and decreasing the preset minimum longitudinal distance threshold; if not, using the preset maximum lateral distance threshold and the preset minimum longitudinal distance threshold.

[0038] Through the embodiments of the present invention, the decision threshold of intersections is dynamically adjusted, which solves the technical problem of insufficient scenario adaptability of traditional methods.

[0039] Figure 3 This is a flowchart illustrating an optional specific implementation method of step S2 in an embodiment of the present invention, as shown below. Figure 3 As shown, the rationality of historical trajectories and the degree of overlap between historical trajectories and lane ranges are used to obtain vulnerable traffic participants after two screenings, including: S21, judging the rationality of historical trajectories based on lateral distance and lateral error between every two adjacent historical trajectory points, and retaining vulnerable traffic participants with reasonable historical trajectories; S22, in response to the overlap between historical trajectories and lane ranges or the current location being an intersection, then retaining vulnerable traffic participants.

[0040] In this embodiment, S22 specifically includes:

[0041] (1) Obtain parameters: Obtain the location of the VRU Get lane width .

[0042] (2) Define the range: Define the lane range and trajectory point range .

[0043] (3) Judgment and clipping: Determine whether the current state is a non-intersection stage, and check whether the trajectory point range of the VRU does not overlap with the lane range. If the condition is not met, filter out the target object. If the condition is met, retain it.

[0044] Through embodiments of the present invention, trajectory rationality assessment filters out VRUs with abrupt or abnormal trajectories (such as sensor noise), reducing the false detection rate. Lane range checking ensures that only VRUs overlapping with the vehicle's path are retained, avoiding the processing of irrelevant targets. Combining intersection conditions enhances adaptability to complex scenarios.

[0045] Based on the above embodiments, the rationality of historical trajectories is determined based on lateral distance and lateral error between every two adjacent historical trajectory points, including: retaining vulnerable traffic participants in response to lateral distance being less than a fifth preset threshold; calculating the weighted sum of lateral errors between every two adjacent historical trajectory points; obtaining the average lateral error based on the weighted sum and the number of historical trajectory points; and retaining vulnerable traffic participants in response to average lateral error being less than a sixth preset threshold.

[0046] In this embodiment, the rationality of the VRU trajectory is determined through the following steps:

[0047] (1) Traverse the set of obstacles and determine whether the current obstacle is a pedestrian or a non-motorized vehicle.

[0048] (2) Obtain trajectory evaluation parameters from the configuration, including: lateral difference weights of trajectory points. Optionally, the closer the horizontal difference is to the current time point, the greater the weight; the number of historical trajectory points Lateral error thresholds in different scenarios (e.g., lane keeping and lane changing scenarios); lateral thresholds at intersections and non-intersections. and .

[0049] (3) Trajectory point acquisition and lateral distance judgment: Acquire the trajectory point of the current VRU at the current time point, including position and direction Determine the lateral distance of the VRU. Does it meet the following conditions:

[0050] (Non-intersection)

[0051] or,

[0052] (Intersection)

[0053] If the instantaneous lateral position of the VRU is already very close to the lane, then retain the VRU; otherwise, filter it out.

[0054] (4) Historical trajectory processing: Check if the historical trajectory of the VRU is empty, and transform the coordinates of the historical trajectory points from the global coordinate system to the Frenet coordinate system: ,in, Distance along the road, This represents the lateral offset.

[0055] (5) Calculation of lateral error and verification of trajectory smoothness:

[0056] from Number of points from start to historical trajectory Traversing all historical trajectory points of the VRU,

[0057] Calculate the lateral error between the current trajectory point and the previous historical trajectory point:

[0058]

[0059] The lateral errors are weighted and accumulated:

[0060]

[0061] Calculate the mean lateral error:

[0062]

[0063] The movement of pedestrians or non-motorized vehicles is continuous and smooth over a short period of time, without drastic changes in position. The question then arises: does the average error exceed a threshold? If the threshold is exceeded, the VRU is filtered out. This method determines whether the trajectory is reasonable by calculating the average lateral error and comparing it with the threshold. If the threshold is exceeded, the object is marked as an anomaly and removed.

[0064] Through the embodiments of the present invention, trajectory smoothness is quantified by weighted average error, abnormal trajectories are accurately identified, false alarms caused by perception errors are reduced, and the reliability of the system in identifying real VRUs is improved.

[0065] Based on the above embodiments, the threat level of vulnerable traffic participants after two screenings is quantified by utilizing their status information. This includes: cropping the historical trajectory of each vulnerable traffic participant after two screenings to obtain a set of trajectory points within a preset time period; and calculating the threat level by using the weighted sum of the reciprocal of the longitudinal distance of the vulnerable traffic participant, the ratio of lateral speed to lateral distance, and the information entropy of the trajectory points in the historical trajectory.

[0066] In this embodiment, trajectory clipping is performed based on a preset time, such as 4 seconds. In an autonomous driving system, the trajectory point set may contain a large number of points, and processing all points will increase the computational burden.

[0067] In this embodiment, the threat quantification model formula is:

[0068]

[0069] Wherein, S is the overall threat score (0~1 normalized value, the higher the score, the greater the threat). These are the weighting coefficients; The longitudinal / lateral distance between the VRU and the vehicle (unit: m); The lateral velocity of the VRU (perpendicular to the lane direction, unit: m / s); The information entropy of the trajectory points reflects the uncertainty of motion.

[0070] By tailoring the trajectory point set according to time constraints through embodiments of the present invention, unnecessary computation can be reduced, improving the efficiency of subsequent behavioral decisions and path planning. The main method is to tailor the time length based on a preset obstacle trajectory; obstacles that pass screening and trajectory evaluation can be used for subsequent behavioral decisions and path planning. Computational resources are allocated through threat level assessment; for example, high-threat targets with S>0.7 are prioritized.

[0071] Figure 2 This is a schematic diagram of a process including zebra crossing screening provided in an embodiment of the present invention, such as... Figure 2 As shown, by utilizing the rationality of historical trajectories and the overlap between historical trajectories and lane ranges, vulnerable traffic participants after a second screening are obtained. Prior to this, the process includes: determining whether the vulnerable traffic participants after the initial screening are located on zebra crossings; if so, the vulnerable traffic participants are taken as actual vulnerable traffic participants; if not, the process of using the rationality of historical trajectories and the overlap between historical trajectories and lane ranges to obtain vulnerable traffic participants after a second screening is performed on the vulnerable traffic participants.

[0072] In this embodiment, zebra crossing parameters are obtained using an OpenDRIVE format map, and a point-in-polygon test is performed using the ray casting method to determine whether the VRU's location is within the zebra crossing. When the VRU is within the detection range of the autonomous driving system and within the zebra crossing boundary, it is filtered as a target object, and the VRU's ID, location information, and trajectory are recorded. When the VRU is within the detection range but not within the zebra crossing boundary, further location and trajectory evaluation is required.

[0073] Through embodiments of the present invention, vulnerable road users can be screened directly at zebra crossings, avoiding system misjudgment of high-risk targets.

[0074] Figure 4 A structural block diagram of the screening system for vulnerable traffic participants provided in an embodiment of the present invention is shown below. Figure 4 As shown, this disclosure provides a screening system for vulnerable traffic participants, which can be used to implement the above-mentioned screening method for vulnerable traffic participants. The system includes: a preliminary screening module, used to obtain vulnerable traffic participants after preliminary screening using their status information; the status information includes historical trajectory and current location; the current location includes the lateral distance and longitudinal distance between the vulnerable traffic participant and the vehicle; a secondary screening module, used to obtain vulnerable traffic participants after secondary screening using the rationality of historical trajectories and the overlap between historical trajectories and lane ranges; and a quantification module, used to quantify the threat level of vulnerable traffic participants after secondary screening using their status information, so that vulnerable traffic participants with a threat level greater than a preset threat level threshold are identified as actual vulnerable traffic participants.

[0075] Based on the same inventive concept, embodiments of the present invention also provide an electronic device. Figure 5 This is a structural block diagram of an electronic device provided in an embodiment of the present invention. Figure 5 As shown, an embodiment of the present invention provides an electronic device including: one or more processors 101, a memory 102, and one or more I / O interfaces 103. The memory 102 stores one or more programs, which, when executed by the one or more processors, enable the one or more processors to implement any of the vulnerable traffic participant screening methods described in the above embodiments; the one or more I / O interfaces 103 are connected between the processors and the memory, configured to enable information interaction between the processors and the memory.

[0076] The processor 101 is a device with data processing capabilities, including but not limited to a central processing unit (CPU); the memory 102 is a device with data storage capabilities, including but not limited to random access memory (RAM, more specifically SDRAM, DDR, etc.), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), and flash memory (FLASH); the I / O interface (read / write interface) 103 is connected between the processor 101 and the memory 102, and can realize information interaction between the processor 101 and the memory 102, including but not limited to a data bus (Bus).

[0077] In some embodiments, the processor 101, memory 102, and I / O interface 103 are interconnected via bus 104, and thus connected to other components of the computing device.

[0078] In some embodiments, the one or more processors 101 include a field-programmable gate array.

[0079] This invention also provides a computer-readable medium. The computer-readable medium stores a computer program, which, when executed by a processor, implements the steps in any of the vulnerable traffic participant screening methods described in the above embodiments. The computer-readable storage medium may be volatile or non-volatile.

[0080] This invention also provides a computer program product, including computer-readable code, or a non-volatile computer-readable storage medium carrying computer-readable code. When the computer-readable code is run in the processor of an electronic device, the processor in the electronic device executes the above-described method for screening vulnerable traffic participants.

[0081] Those skilled in the art will understand that all or some of the steps, systems, and apparatuses disclosed above, and their functional modules / units, can be implemented as software, firmware, hardware, or suitable combinations thereof. In hardware implementations, the division between functional modules / units mentioned above does not necessarily correspond to the division of physical components; for example, a physical component may have multiple functions, or a function or step may be performed collaboratively by several physical components. Some or all physical components may be implemented as software executed by a processor, such as a central processing unit, digital signal processor, or microprocessor, or as hardware, or as an integrated circuit, such as an application-specific integrated circuit (ASIC). Such software can be distributed on a computer-readable storage medium, which may include computer storage media (or non-transitory media) and communication media (or transient media).

[0082] As is known to those skilled in the art, the term computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information, such as computer-readable program instructions, data structures, program modules, or other data. Computer storage media includes, but is not limited to, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), static random access memory (SRAM), flash memory or other memory technologies, portable compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical disc storage, magnetic cartridges, magnetic tape, disk storage or other magnetic storage devices, or any other medium that can be used to store desired information and is accessible to a computer. Furthermore, it is known to those skilled in the art that communication media typically contain computer-readable program instructions, data structures, program modules, or other data in modulated data signals such as carrier waves or other transmission mechanisms, and may include any information delivery medium.

[0083] The computer-readable program instructions described herein can be downloaded from computer-readable storage media to various computing / processing devices, or downloaded via a network, such as the Internet, local area network, wide area network, and / or wireless network, to an external computer or external storage device. The network may include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. A network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards them to the computer-readable storage media in the respective computing / processing device.

[0084] The computer program instructions used to perform the operations of this invention may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk, C++, etc., and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The computer-readable program instructions may be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or may be connected to an external computer (e.g., via the Internet using an Internet service provider). In some embodiments, electronic circuitry, such as programmable logic circuitry, field-programmable gate arrays (FPGAs), or programmable logic arrays (PLAs), is personalized by utilizing state information from the computer-readable program instructions. This electronic circuitry can execute the computer-readable program instructions to implement various aspects of the invention.

[0085] The computer program product described herein can be implemented specifically through hardware, software, or a combination thereof. In one alternative embodiment, the computer program product is specifically embodied in a computer storage medium; in another alternative embodiment, the computer program product is specifically embodied in a software product, such as a software development kit (SDK), etc.

[0086] Various aspects of the present invention are described herein with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It should 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-readable program instructions.

[0087] These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that, when executed by the processor of the computer or other programmable data processing apparatus, they create means for implementing the functions / actions specified in one or more blocks of the flowchart and / or block diagram. These computer-readable program instructions can also be stored in a computer-readable storage medium that causes a computer, programmable data processing apparatus, and / or other device to operate in a particular manner; thus, the computer-readable medium storing the instructions comprises an article of manufacture that includes instructions for implementing aspects of the functions / actions specified in one or more blocks of the flowchart and / or block diagram.

[0088] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions executed on the computer, other programmable data processing apparatus, or other device to perform the functions / actions specified in one or more boxes of a flowchart and / or block diagram.

[0089] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of an instruction, which contains one or more executable instructions for implementing a specified logical function. In some alternative implementations, the functions marked in the blocks may occur in a different order than those shown in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0090] Example embodiments have been disclosed herein, and while specific terminology has been used, it is for illustrative purposes only and should be construed as such, and is not intended to be limiting. In some instances, it will be apparent to those skilled in the art that features, characteristics, and / or elements described in conjunction with particular embodiments may be used alone, or in combination with features, characteristics, and / or elements described in conjunction with other embodiments, unless otherwise expressly indicated. Therefore, those skilled in the art will understand that various changes in form and detail may be made without departing from the scope of the invention as set forth in the appended claims.

Claims

1. A method for screening vulnerable traffic participants, characterized in that, include: The status information of vulnerable traffic participants is used to obtain the initially screened vulnerable traffic participants; The status information includes historical trajectory and current location; the current location includes the lateral and longitudinal distances between vulnerable road users and vehicles; By utilizing the rationality of historical trajectories and the degree of overlap between historical trajectories and lane ranges, vulnerable traffic participants are identified after a second screening. By utilizing the status information of vulnerable traffic participants, the threat level of the vulnerable traffic participants after the second screening is quantified so that vulnerable traffic participants with a threat level greater than a preset threat level threshold are identified as actual vulnerable traffic participants.

2. The method according to claim 1, wherein, The process of obtaining initially screened vulnerable traffic participants using their status information includes: By utilizing the historical trajectories of vulnerable traffic participants, the time points at which their movement direction changed and their movement direction were extracted. In response to the existence of a vulnerable traffic participant that meets the first distance condition, the vulnerable traffic participant is filtered out; the distance condition is that the longitudinal distance is less than a preset minimum longitudinal distance threshold, or the lateral distance is greater than a preset maximum lateral distance threshold, or the longitudinal distance is greater than a first preset threshold and the time for the longitudinal distance to be greater than the first preset threshold is greater than a second preset threshold. If a vulnerable traffic participant meets the second distance condition among the filtered vulnerable traffic participants, then the vulnerable traffic participant is retained; the second distance condition is that the lateral distance is less than the third preset threshold, or the difference between the current direction of movement and the historical average direction of movement is less than the fourth preset threshold. The longitudinal direction is the direction along the tangent of the lane centerline; the transverse direction is the direction perpendicular to the lane centerline.

3. The method according to claim 2, wherein, The method of extracting the time points and directions of change of movement by utilizing the historical trajectories of vulnerable traffic participants also includes: Adjusting preset configuration parameters specifically includes: the preset configuration parameters include a preset maximum horizontal distance threshold and a preset minimum vertical distance threshold; Determine if the current location is an intersection; If so, increase the preset maximum horizontal distance threshold and decrease the preset minimum vertical distance threshold; If not, then the preset maximum horizontal distance threshold and the preset minimum vertical distance threshold will be used.

4. The method according to claim 1, wherein, The process of utilizing the rationality of historical trajectories and the overlap between historical trajectories and lane ranges to identify vulnerable traffic participants after the second screening includes: The rationality of historical trajectories is judged based on lateral distance and lateral error between every two adjacent historical trajectory points, and vulnerable traffic participants with reasonable historical trajectories are retained. If the historical trajectory overlaps with the lane range or the current location is an intersection, the vulnerable traffic participant is retained.

5. The method according to claim 4, wherein, The determination of the rationality of historical trajectories based on lateral distance and the lateral error between every two adjacent historical trajectory points includes: If the lateral distance is less than a fifth preset threshold, the vulnerable traffic participant is retained. Calculate the weighted sum of the lateral errors between every two adjacent historical trajectory points; The average lateral error is obtained based on the weighted sum and the number of historical trajectory points. If the average lateral error is less than a sixth preset threshold, the vulnerable traffic participant is retained.

6. The method according to claim 1, wherein, The method of quantifying the threat level of vulnerable traffic participants after the second screening by utilizing their status information includes: The historical trajectory of each disadvantaged traffic participant after the second screening is cut off to obtain a set of trajectory points within a preset time period; The threat level is calculated by weighting the reciprocal of the longitudinal distance of vulnerable traffic participants, the ratio of lateral velocity to lateral distance, and the entropy of trajectory points in historical trajectories.

7. The method according to claim 1, wherein, The process of using the rationality of historical trajectories and the overlap between historical trajectories and lane ranges to obtain vulnerable traffic participants after a second screening includes: Determine whether the vulnerable traffic participants identified in the initial screening are located on the zebra crossing; If so, then vulnerable traffic participants will be considered as actual vulnerable traffic participants; If not, then the steps of using historical trajectories to determine the rationality of vulnerable traffic participants and the degree of overlap between historical trajectories and lane ranges are performed to obtain vulnerable traffic participants after a second screening.

8. A screening system for vulnerable traffic participants, characterized in that, include: The initial screening module is used to obtain the initial screening of vulnerable traffic participants by utilizing their status information; the status information includes historical trajectory and current location; the current location includes the lateral distance and longitudinal distance between the vulnerable traffic participant and the vehicle; The second screening module is used to identify vulnerable traffic participants by utilizing the rationality of historical trajectories and the degree of overlap between historical trajectories and lane ranges. The quantification module is used to quantify the threat level of the vulnerable traffic participants after the second screening by utilizing the status information of the vulnerable traffic participants, so as to identify the vulnerable traffic participants whose threat level is greater than a preset threat level threshold as the actual vulnerable traffic participants.

9. An electronic device, characterized in that, include: One or more processors; Memory, used to store one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the method as described in any one of claims 1 to 7.

10. A computer-readable medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method as described in any one of claims 1 to 7.

Images