A method and device for determining a pedestrian motion trajectory

Abstract

The application discloses a kind of pedestrian motion trajectory determination method and determination device, the pedestrian motion trajectory determination method is used in computer vision technical field, specifically includes: obtaining the current scene where the pedestrian is located;When the current scene is first scene, the motion information of the pedestrian is input to first model to obtain the first motion trajectory of the pedestrian;When the current scene is second scene, the motion information of the pedestrian and other traffic participants is input to second model to obtain the second motion trajectory of the pedestrian.The application combines the characteristics of high-precision map, simplifies the pedestrian motion prediction under the pedestrian crossing scene;Meanwhile, the interaction of pedestrian and other obstacles is considered in other scenes, the historical motion characteristics of pedestrian are extracted using deep learning model, pedestrian trajectory learning modeling is carried out on large-scale data set, the accuracy of pedestrian prediction result is improved, and the robustness and real-time performance of model prediction are improved.

Description

Technical Field

[0001] This disclosure relates to the field of computer vision technology, and more specifically, to a method and apparatus for determining the trajectory of a pedestrian. Background Technology

[0002] With the continuous advancement of technology, various self-driving cars, intelligent robots, intelligent wheelchairs, and other intelligent terminals have emerged. During the movement of these intelligent terminals, obstacle avoidance maneuvers are performed by predicting the movement trajectories of pedestrians ahead. Therefore, accurately predicting pedestrian movement trajectories is the core issue for autonomous obstacle avoidance by intelligent terminals.

[0003] Currently, pedestrian trajectory prediction is mainly applied in the fields of service robots, autonomous driving, and video surveillance. Among the research methods in these fields, some consider the prediction of individual pedestrians, some consider the prediction of interactions between pedestrians, and some are based on static environmental factors. However, there is no method that comprehensively considers both individual pedestrians and interactions between pedestrians. Furthermore, the accuracy of the motion trajectory predicted by existing methods is low, and the effectiveness of motion trajectory prediction is poor. Summary of the Invention

[0004] This disclosure provides a method and apparatus for determining pedestrian movement trajectories, which at least solves the technical problems of low accuracy and poor effectiveness of existing prediction methods in predicting movement trajectories.

[0005] According to one aspect of the present disclosure, a method for determining a pedestrian's movement trajectory is provided, comprising: acquiring the current scene in which the pedestrian is located; when the current scene is a first scene, inputting the pedestrian's movement information into a first model to obtain a first movement trajectory of the pedestrian; and when the current scene is a second scene, inputting the movement information of the pedestrian and other traffic participants into a second model to obtain a second movement trajectory of the pedestrian.

[0006] In an exemplary embodiment, obtaining the current scene where the pedestrian is located includes:

[0007] Determine whether map information can be obtained at the current moment; if map information can be obtained, determine whether the pedestrian's current location is within a predetermined area; if the pedestrian's current location is within the predetermined area, the current scenario is the first scenario; if the pedestrian's current location is outside the predetermined area, the current scenario is the second scenario.

[0008] In one exemplary embodiment, the method further includes: if the map information cannot be obtained, the current scenario is a second scenario.

[0009] In an exemplary embodiment, inputting the pedestrian's motion information into a first model to obtain the pedestrian's first motion trajectory includes: obtaining the pedestrian's current state information and current motion control information, wherein the current state information includes at least position information and velocity information, and the current motion control information includes at least acceleration and angular velocity; establishing a kinematic model of the pedestrian based on the current state information and the current motion control information; and obtaining the pedestrian's first motion trajectory through the kinematic model.

[0010] In an exemplary embodiment, inputting the motion information of the pedestrian and other traffic participants into the second model to obtain the second motion trajectory of the pedestrian includes:

[0011] The system acquires the current location information of the pedestrian and other traffic participants; extracts the historical trajectory information of the pedestrian and other traffic participants through an extraction model; based on the current location information and the historical trajectory information, it acquires the interaction relationship information of the pedestrian with other traffic participants and obstacles through an interaction model; and acquires the second movement trajectory of the pedestrian based on the interaction relationship information.

[0012] In one exemplary embodiment, before obtaining the current scene in which the pedestrian is located, the method further includes: obtaining the state information of the obstacle.

[0013] In one exemplary embodiment, the state information of the obstacle includes at least the obstacle's historical location, type, and shape information.

[0014] Secondly, embodiments of this disclosure also provide a device for determining the trajectory of a pedestrian, comprising:

[0015] The first acquisition module is used to acquire the current scene where the pedestrian is located; the second acquisition module is used to input the pedestrian's motion information into the first model to acquire the pedestrian's first motion trajectory when the current scene is the first scene; the third acquisition module is used to input the motion information of the pedestrian and other traffic participants into the second model to acquire the pedestrian's second motion trajectory when the current scene is the second scene.

[0016] Thirdly, embodiments of this disclosure also provide a computer-readable storage medium storing a computer program for executing the determination method described in any of the above technical solutions.

[0017] Fourthly, embodiments of this disclosure also provide an electronic device, the electronic device comprising: a processor and a memory for storing processor-executable instructions; wherein the processor is configured to execute the determination method described in any of the above technical solutions.

[0018] As can be seen from the above, the pedestrian trajectory determination method provided in this disclosure combines the features of high-precision maps to simplify pedestrian movement prediction in crosswalk scenarios; at the same time, it considers the interaction between pedestrians and other obstacles in other scenarios, uses a deep learning model to extract historical movement features of pedestrians, and performs pedestrian trajectory learning and modeling on a large-scale dataset, thereby improving the accuracy of pedestrian prediction results. Through this multimodal pedestrian trajectory determination method, the robustness and real-time performance of model prediction are improved without sacrificing pedestrian prediction accuracy.

[0019] To make the above-mentioned objects, features and advantages of this disclosure more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of this disclosure or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this disclosure. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 This is a flowchart of the steps of the pedestrian trajectory prediction method provided in this disclosure;

[0022] Figure 2 This is a flowchart of the steps for obtaining the current scene of the pedestrian provided in this disclosure;

[0023] Figure 3 This is a flowchart of the steps for obtaining the first motion trajectory of the pedestrian provided in this disclosure;

[0024] Figure 4 This is a flowchart of the steps for obtaining the second motion trajectory of the pedestrian provided in this disclosure;

[0025] Figure 5 This is a design diagram of predicting pedestrian trajectories using neural networks, provided in this disclosure;

[0026] Figure 6 This is a structural block diagram of the pedestrian trajectory determination device provided in this disclosure;

[0027] Figure 7 This is a structural block diagram of the electronic device provided in this disclosure. Detailed Implementation

[0028] The specific embodiments of this disclosure will now be described in detail with reference to the accompanying drawings, but these are not intended to limit the scope of this disclosure.

[0029] It should be understood that various modifications can be made to the embodiments disclosed herein. Therefore, the above description should not be considered as limiting, but merely as an example of embodiments. Other modifications within the scope and spirit of this disclosure will be apparent to those skilled in the art.

[0030] The accompanying drawings, which are included in and form part of this specification, illustrate embodiments of the present disclosure and, together with the general description of the disclosure given above and the detailed description of the embodiments given below, serve to explain the principles of the disclosure.

[0031] These and other features of this disclosure will become apparent from the following description of preferred forms of embodiments given as non-limiting examples, with reference to the accompanying drawings.

[0032] It should also be understood that although this disclosure has been described with reference to specific examples, many other equivalent forms that can be definitively implemented by those skilled in the art also fall within the scope of protection defined herein.

[0033] The above and other aspects, features and advantages of this disclosure will become more apparent when taken in conjunction with the accompanying drawings and in view of the following detailed description.

[0034] This specification may use the phrases “in one embodiment,” “in another embodiment,” “in yet another embodiment,” or “in still another embodiment,” all of which may refer to one or more of the same or different embodiments according to this disclosure.

[0035] The present disclosure will be further described below with reference to the accompanying drawings and specific embodiments.

[0036] Example 1

[0037] This disclosure relates to the field of computer vision technology, specifically to a method for determining the trajectory of a pedestrian. The method for determining the trajectory can be applied to a smart terminal, which is a mobile smart terminal, such as a robot, a smart wheelchair, or a self-driving car.

[0038] like Figure 1 As shown, the prediction method described in this disclosure includes steps S101 to S103. The specific implementation principles of each step are as follows:

[0039] S101, Obtain the current scene where the pedestrian is located.

[0040] In this step, the current scene in which the pedestrian is located is obtained. The scene in which the pedestrian is located includes a scene in which the pedestrian's goal is relatively clear, such as when the pedestrian is on a crosswalk or a sidewalk. It can also include a scene in which the pedestrian is present with other traffic participants, the pedestrian's goal is not clear, and the pedestrian is in a vulnerable position among the traffic participants, such as when the pedestrian is not on a crosswalk or on a rural road.

[0041] like Figure 2 As shown, the specific steps for obtaining the current scene of the pedestrian provided in this disclosure include the following:

[0042] S201, determine whether map information can be obtained at the current moment.

[0043] In this step, it is determined whether map information can be obtained at the current moment, and the location of the pedestrian can be determined through the map information.

[0044] S202, if map information is available, determine whether the pedestrian's current location is within a predetermined area.

[0045] After completing step S201 above, in this step, if map information can be obtained, it is determined whether the current location of the pedestrian is within a predetermined area, wherein the predetermined area is a scenario where the pedestrian's target is relatively clear, such as the situation where the pedestrian is on the sidewalk.

[0046] Specifically, the smart terminal can obtain map information from the server and determine the current location of the pedestrian based on the map information. The current location can be a scene where the pedestrian's target is relatively clear, such as a sidewalk, or a scene where pedestrians and other traffic participants coexist, such as a non-sidewalk.

[0047] S203, if the pedestrian's current position is within the predetermined area, the current scenario is the first scenario.

[0048] After completing step S202 above, in this step, when map information is obtained, if the pedestrian's current position is in a scenario where the pedestrian's target is relatively clear, such as on a sidewalk, the current scenario is the first scenario; if the pedestrian's current position is outside the predetermined area, the current scenario is the second scenario.

[0049] Specifically, in the second scenario, pedestrians and other traffic participants coexist. The pedestrians' goals are not clear and they are in a vulnerable position among the traffic participants, such as when they are on non-pedestrian walkways or on rural roads. In this case, it is necessary to consider the movement interaction between people and other obstacles to avoid collisions.

[0050] S102, when the current scene is the first scene, the pedestrian's motion information is input into the first model to obtain the pedestrian's first motion trajectory.

[0051] After completing step S101 above, in this step, when the current scene is the first scene, the pedestrian's motion information is input into the first model to obtain the pedestrian's first motion trajectory.

[0052] Specifically, considering that pedestrians are in scenarios where the target is relatively clear, such as sidewalks, the research adopts a target-oriented approach. In this disclosure, the unicycle model is used to predict the movement trajectory of pedestrians.

[0053] like Figure 3 As shown, obtaining the pedestrian's first movement trajectory specifically includes the following steps:

[0054] S301, Obtain the current state information and current motion control information of the pedestrian.

[0055] In this step, the current state information and current motion control information of the pedestrian are obtained, wherein the current state information includes at least position information and velocity information, and the current motion control information includes at least acceleration and angular velocity.

[0056] Specifically, the pedestrian's state is denoted as X∈R 4 The control information is labeled u∈R 2 Simultaneously, noise is processed as w ~ N(0, W) ∈ R. 4 The detailed definitions are as follows:

[0057]

[0058] Here, x and y are labeled as the spatial coordinates of the pedestrian, velocity is labeled as v, velocity direction is labeled as θ, and acceleration and angular velocity are labeled as a and ω, respectively. These labels can represent the physical state at a specified moment.

[0059] S302, establish the kinematic model of the pedestrian based on the current state information of S and the current motion control information.

[0060] After completing step S301 above, in this step, a kinematic model of the pedestrian is established based on the current state information and the current motion control information.

[0061] Specifically, based on the pedestrian's current state information and current motion control information, a kinematic model of the pedestrian is established, as follows:

[0062]

[0063] S303, the first motion trajectory of the pedestrian is obtained through the kinematic model.

[0064] After obtaining the kinematic model of the pedestrian, in this step, the first motion trajectory of the pedestrian is obtained based on the kinematic model.

[0065] Specifically, based on the pedestrian's current position, speed, and angle, the system calculates the pedestrian's state information for the next moment; and iterates further to obtain all state information for the required predicted length. By using a kinematic model to model pedestrians, the motion prediction process for pedestrians in crosswalk scenarios is simplified, avoiding the performance overhead of using deep learning models to predict crowds.

[0066] S103, when the current scenario is the second scenario, the motion information of the pedestrian and other traffic participants is input into the second model to obtain the second motion trajectory of the pedestrian.

[0067] In this step, when the smart terminal determines that the current scene is the second scene, it inputs the pedestrian's motion information into the second model to obtain the pedestrian's second motion trajectory.

[0068] Specifically, considering that pedestrians are in scenarios other than crosswalks or in areas not covered by high-precision maps, their goals are not clear and they are at a disadvantage among traffic participants. To predict the movement of these pedestrians, it is necessary to consider the movement interaction between pedestrians and other obstacles to avoid collisions. Therefore, the state information of the obstacles should be obtained in advance. The state information of the obstacles includes at least the historical location, type, and shape information of the obstacles.

[0069] like Figure 4 As shown, obtaining the pedestrian's second movement trajectory specifically includes the following steps:

[0070] S401, Obtain the current location information of the pedestrians and other traffic participants.

[0071] In this step, the current location information of the pedestrians and other traffic participants is obtained.

[0072] S402, obtain the historical trajectory information of the pedestrians and other traffic participants by extracting the model.

[0073] In this step, historical trajectory information of the pedestrians and other traffic participants is obtained by extracting a model. Specifically, such as... Figure 5As shown, a deep learning model is used to model the historical observation trajectory of a pedestrian. The output of the deep learning network is a temporal feature map, consisting of three sets of one-dimensional convolutions, each with two residual blocks, and the stride of the first block is 2. Then, a Feature Pyramid Network (FPN) is used to fuse multi-scale features, and another residual block is used to obtain the output tensor. All convolutional kernels have a size of 3, the number of output channels is 128, and layer normalization and ReLU operations are performed after each convolutional operation. The historical trajectory of the pedestrian is extracted using CNN and FPN networks, and the historical observation trajectory of the pedestrian is represented as {Δp}. -(T-1) , ..., Δp -1 , Δp0}, where Δp t The input tensor represents the two-dimensional coordinate information from time t-1 to time t, where T is the length of the observed trajectory. If the observation length is less than T, it is padded with 0s. A 1×T label is added to indicate whether the trajectory coordinates are padded, resulting in an input tensor size of 3×T. Then, a 1D CNN is used to process the trajectory input to improve its effectiveness in extracting multi-scale features and the efficiency of parallel computation.

[0074] S403, based on the current location information and the historical trajectory information, obtain the interaction relationship information of the pedestrian with other traffic participants and obstacles through the interaction model.

[0075] After completing steps S401 and S402 above, in this step, based on the current location information and the historical trajectory information, the interaction relationship information between the pedestrian and other traffic participants and obstacles is obtained through the interaction model.

[0076] Specifically, after the observed trajectory information features of each pedestrian and traffic participant are extracted by the network, another network is used to process the interaction relationships of the traffic participants. This network consists of two residual blocks, each containing an attention layer, a linear layer, and residual connections, with each layer having 128 output channels. The attention layer is used to characterize the interaction relationships of traffic participants. Given participant node i, the aggregated features with another participant j are defined as follows:

[0077]

[0078] Where x i Let represent the features of node i, W be the weight matrix, Ψ be the combination of the normalization layer and the ReLU operation, and Δ be the weight matrix. i，j =MLP(v j -v i ), where v represents the coordinate information of the node.

[0079] S404, Obtain the pedestrian's second movement trajectory based on the interaction relationship information.

[0080] After completing step S403 above, in this step, after obtaining the 128-dimensional output channel through the two networks mentioned above, a prediction head is added to obtain the predicted trajectory.

[0081] The pedestrian trajectory determination method provided in this disclosure combines the features of high-precision maps to simplify pedestrian movement prediction in crosswalk scenarios. At the same time, it considers the interaction between pedestrians and other obstacles in other scenarios, uses a deep learning model to extract historical movement features of pedestrians, and performs pedestrian trajectory learning and modeling on a large-scale dataset, thereby improving the accuracy of pedestrian prediction results. Through this multimodal pedestrian trajectory determination method, the robustness and real-time performance of model prediction are improved without sacrificing pedestrian prediction accuracy.

[0082] Example 2

[0083] To better implement the above methods, a second aspect of this disclosure also provides a pedestrian trajectory determination device that can be integrated into an electronic device.

[0084] For example, such as Figure 6 As shown, the pedestrian trajectory determination device 200 may include: a first acquisition module 210, a second acquisition module 220, and a third acquisition module 230, as detailed below:

[0085] (1) The first acquisition module 210 is used to acquire the current scene where the pedestrian is located;

[0086] (2) The second acquisition module 220 is used to input the pedestrian's motion information into the first model to obtain the pedestrian's first motion trajectory when the current scene is the first scene;

[0087] The third acquisition module 230 is used to input the motion information of the pedestrian and other traffic participants into the second model to obtain the second motion trajectory of the pedestrian when the current scene is the second scene.

[0088] The pedestrian trajectory determination device provided in this embodiment combines the features of high-precision maps to simplify pedestrian movement prediction in crosswalk scenarios. At the same time, it considers the interaction between pedestrians and other obstacles in other scenarios, uses a deep learning model to extract historical movement features of pedestrians, and performs pedestrian trajectory learning and modeling on a large-scale dataset, thereby improving the accuracy of pedestrian prediction results. Through this multimodal pedestrian trajectory determination method, the robustness and real-time performance of the model prediction are improved without sacrificing the accuracy of pedestrian prediction.

[0089] Example 3

[0090] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be performed by instructions, or by instructions controlling related hardware. These instructions can be stored in a computer-readable storage medium and loaded and executed by a processor.

[0091] Therefore, a third embodiment of this disclosure provides a storage medium, which is a computer-readable medium storing a computer program. When executed by a processor, the computer program implements the motion trajectory determination method provided in this embodiment of the disclosure, including the following steps S11 to S13:

[0092] S11, Obtain the current scene where the pedestrian is located;

[0093] S12, when the current scene is the first scene, the pedestrian's motion information is input into the first model to obtain the pedestrian's first motion trajectory;

[0094] S13, when the current scenario is the second scenario, the motion information of the pedestrian and other traffic participants is input into the second model to obtain the second motion trajectory of the pedestrian.

[0095] Furthermore, when the computer program is executed by a processor, it implements other methods provided in any of the above embodiments of this disclosure.

[0096] The pedestrian trajectory determination method provided in this disclosure combines the features of high-precision maps to simplify pedestrian movement prediction in crosswalk scenarios. At the same time, it considers the interaction between pedestrians and other obstacles in other scenarios, uses a deep learning model to extract historical movement features of pedestrians, and performs pedestrian trajectory learning and modeling on a large-scale dataset, thereby improving the accuracy of pedestrian prediction results. Through this multimodal pedestrian trajectory determination method, the robustness and real-time performance of model prediction are improved without sacrificing pedestrian prediction accuracy.

[0097] Example 4

[0098] The fourth embodiment of this disclosure provides an electronic device, such as... Figure 7 As shown, the electronic device includes at least a processor 401 and a memory 402. The memory 402 stores a computer program, and the processor 401 implements the pedestrian trajectory determination method provided in any embodiment of this disclosure when executing the computer program in the memory 402. For example, the method executed by the computer program in the electronic device is as follows:

[0099] S21, Obtain the current scene where the pedestrian is located;

[0100] S22, when the current scene is the first scene, the pedestrian's motion information is input into the first model to obtain the pedestrian's first motion trajectory;

[0101] S23, when the current scenario is the second scenario, the motion information of the pedestrian and other traffic participants is input into the second model to obtain the second motion trajectory of the pedestrian.

[0102] In specific implementation, the first acquisition module 210, the second acquisition module 220, and the determination module 230 are all stored as program units in the memory 402, and the processor 401 executes the program units stored in the memory 402 to implement the corresponding functions.

[0103] The pedestrian trajectory determination method provided in this disclosure combines the features of high-precision maps to simplify pedestrian movement prediction in crosswalk scenarios. At the same time, it considers the interaction between pedestrians and other obstacles in other scenarios, uses a deep learning model to extract historical movement features of pedestrians, and performs pedestrian trajectory learning and modeling on a large-scale dataset, thereby improving the accuracy of pedestrian prediction results. Through this multimodal pedestrian trajectory determination method, the robustness and real-time performance of model prediction are improved without sacrificing pedestrian prediction accuracy.

[0104] The aforementioned storage medium may be included in the aforementioned electronic device; or it may exist independently and not be assembled into the electronic device.

[0105] The aforementioned storage medium carries one or more programs, which, when executed by the electronic device, cause the electronic device to: acquire at least two Internet Protocol (IP) addresses; send a node evaluation request, including at least two IP addresses, to a node evaluation device, wherein the node evaluation device selects an IP address from the at least two IP addresses and returns it; and receive the IP address returned by the node evaluation device; wherein the acquired IP address indicates an edge node in the content delivery network.

[0106] Alternatively, the storage medium may carry one or more programs that, when executed by the electronic device, cause the electronic device to: receive a node evaluation request including at least two Internet Protocol (IP) addresses; select an IP address from the at least two IP addresses; and return the selected IP address; wherein the received IP address indicates an edge node in the content delivery network.

[0107] Computer program code for performing the operations of this disclosure can be written in one or more programming languages ​​or a combination thereof, including but not limited to object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the passenger's computer, partially on the passenger's computer, as a standalone software package, partially on the passenger'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 can be connected to the passenger's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0108] It should be noted that the storage medium described in this disclosure can be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this disclosure, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In this disclosure, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium may also be any storage medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the storage medium can be transmitted using any suitable medium, including but not limited to: wires, optical fibers, RF (radio frequency), etc., or any suitable combination thereof.

[0109] 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 this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated 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, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0110] The units described in the embodiments of this disclosure can be implemented in software or hardware. The names of the units are not, in some cases, intended to limit the specific unit.

[0111] The functions described above in this document can be performed, at least in part, by one or more hardware logic components. For example, exemplary types of hardware logic components that can be used, without limitation, include: Field Programmable Gate Arrays (FPGAs), Application-Specific Integrated Circuits (ASICs), Application Standard Products (ASSPs), System-on-Chip (SoCs), Complex Programmable Logic Devices (CPLDs), and so on.

[0112] In the context of this disclosure, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0113] The above description is merely a preferred embodiment of this disclosure and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of this disclosure is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-described concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features disclosed in this disclosure that have similar functions.

[0114] Furthermore, while the operations are described in a specific order, this should not be construed as requiring these operations to be performed in the specific order shown or in a sequential order. In certain environments, multitasking and parallel processing may be advantageous. Similarly, while several specific implementation details are included in the above discussion, these should not be construed as limiting the scope of this disclosure. Certain features described in the context of individual embodiments may also be implemented in combination in a single embodiment. Conversely, various features described in the context of a single embodiment may also be implemented individually or in any suitable sub-combination in multiple embodiments.

[0115] The foregoing has provided a detailed description of several embodiments of this disclosure. However, this disclosure is not limited to these specific embodiments. Those skilled in the art can make various variations and modifications based on the concept of this disclosure, and all such variations and modifications should fall within the scope of protection claimed by this disclosure.

Claims

1. A method for determining the trajectory of a pedestrian, characterized in that, include: Determine whether map information can be obtained at the current moment; If map information is available, determine whether the pedestrian's current location is within the predetermined area; If the pedestrian's current location is within the predetermined area, the current scenario in which the pedestrian is located is determined to be the first scenario; If the pedestrian's current location is outside the predetermined area, the current scenario is determined to be the second scenario; When the current scenario is the first scenario, the current state information and current motion control information of the pedestrian are obtained. The current state information includes at least position information and velocity information, and the current motion control information includes at least acceleration and angular velocity. A kinematic model of the pedestrian is established based on the current state information and the current motion control information; The first motion trajectory of the pedestrian is obtained through the kinematic model. When the current scenario is the second scenario, obtain the current location information of the pedestrians and other traffic participants; Historical trajectory information of the pedestrian and other traffic participants is obtained by extracting a model; based on the current location information and the historical trajectory information, the interaction relationship information of the pedestrian with other traffic participants and obstacles is obtained by an interaction model. The pedestrian's second movement trajectory is obtained based on the interaction relationship information.

2. The determination method according to claim 1, characterized in that, Also includes: If the map information cannot be obtained, the current scenario is the second scenario.

3. The determination method according to claim 1, characterized in that, Before obtaining the current scene where the pedestrian is located, the method further includes: obtaining the state information of the obstacle.

4. The determination method according to claim 1, characterized in that, The status information of the obstacle includes at least its historical location, type, and shape information.

5. A device for determining the trajectory of a pedestrian, characterized in that, include: The first acquisition module is used to determine whether map information can be acquired at the current moment; If map information is available, determine whether the pedestrian's current location is within the predetermined area; If the pedestrian's current location is within the predetermined area, the current scenario where the pedestrian is located is determined to be the first scenario; if the pedestrian's current location is outside the predetermined area, the current scenario is determined to be the second scenario. The second acquisition module is used to acquire the current state information and current motion control information of the pedestrian when the current scene is the first scene. The current state information includes at least position information and velocity information, and the current motion control information includes at least acceleration and angular velocity. A kinematic model of the pedestrian is established based on the current state information and the current motion control information; The first motion trajectory of the pedestrian is obtained through the kinematic model. The third acquisition module is used to acquire the current location information of the pedestrians and other traffic participants when the current scenario is the second scenario. Historical trajectory information of the pedestrian and other traffic participants is obtained by extracting a model; based on the current location information and the historical trajectory information, the interaction relationship information of the pedestrian with other traffic participants and obstacles is obtained by an interaction model. The pedestrian's second movement trajectory is obtained based on the interaction relationship information.

6. A computer-readable storage medium storing a computer program for performing the determination method according to any one of claims 1-4.

7. An electronic device, the electronic device comprising: processor; Memory used to store the processor's executable instructions; The processor is configured to execute the determination method according to any one of claims 1-4.

Images