CN115817531A - Pedestrian motion track determination method and determination device - Google Patents
Pedestrian motion track determination method and determination device Download PDFInfo
- Publication number
- CN115817531A CN115817531A CN202211502611.7A CN202211502611A CN115817531A CN 115817531 A CN115817531 A CN 115817531A CN 202211502611 A CN202211502611 A CN 202211502611A CN 115817531 A CN115817531 A CN 115817531A
- Authority
- CN
- China
- Prior art keywords
- pedestrian
- motion
- information
- scene
- current
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Abstract
The invention discloses a method and a device for determining a pedestrian motion trail, wherein the method for determining the pedestrian motion trail is used in the technical field of computer vision and specifically comprises the following steps: acquiring a current scene where the pedestrian is located; when the current scene is a first scene, inputting the motion information of the pedestrian to a first model to obtain a first motion track of the pedestrian; and when the current scene is a second scene, inputting the motion information of the pedestrian and other traffic participants into a second model to obtain a second motion track of the pedestrian. The method combines the characteristics of the high-precision map, and simplifies the pedestrian motion prediction in the pedestrian crossing scene; meanwhile, interaction of pedestrians and other obstacles is considered in other scenes, historical motion characteristics of the pedestrians are extracted by using a deep learning model, and pedestrian trajectory learning modeling is carried out on a large-scale data set, so that the accuracy of a pedestrian prediction result is improved, and the robustness and the real-time performance of model prediction are improved.
Description
Technical Field
The disclosure relates to the technical field of computer vision, in particular to a method and a device for determining a pedestrian motion track.
Background
With the continuous progress of science and technology, various intelligent terminals such as unmanned automobiles, intelligent robots, intelligent wheelchairs and the like appear. In the moving process of the intelligent terminal, the obstacle avoidance operation is carried out by predicting the motion trail of the pedestrian in front. Therefore, accurate presetting of the pedestrian movement track is the core problem of autonomous obstacle avoidance of the intelligent terminal.
At present, pedestrian trajectory prediction is mainly applied to the fields of service robots, automatic driving and video monitoring, and in research methods in the fields, single pedestrian prediction is considered, inter-pedestrian interaction prediction is considered, and prediction is performed based on static environment factors, but the inter-pedestrian interaction between a single pedestrian and a pedestrian is not considered comprehensively, and the accuracy of a motion trajectory predicted by an existing prediction mode is low, and the effectiveness of motion trajectory prediction is poor.
Disclosure of Invention
The embodiment of the disclosure provides a method and a device for determining a pedestrian motion trail, so as to at least solve the technical problems of low accuracy of the motion trail predicted by the existing prediction mode and poor effectiveness of motion trail prediction.
According to an aspect of the embodiments of the present disclosure, there is provided a method for determining a pedestrian motion trajectory, including: acquiring a current scene where the pedestrian is located; when the current scene is a first scene, inputting the motion information of the pedestrian to a first model to obtain a first motion track of the pedestrian; and when the current scene is a second scene, inputting the motion information of the pedestrian and other traffic participants into a second model to obtain a second motion track of the pedestrian.
In an exemplary embodiment, the acquiring the current scene of the pedestrian includes:
judging whether the map information can be acquired at the current moment; under the condition that map information can be acquired, judging whether the current position of the pedestrian is located in a preset area; under the condition that the current position of the pedestrian is located in the predetermined area, the current scene is a first scene; and under the condition that the current position of the pedestrian is located outside the preset area, the current scene is a second scene.
In one exemplary embodiment, further comprising: and in the case of failing to acquire the map information, the current scene is a second scene.
In an exemplary embodiment, the inputting the motion information of the pedestrian to the first model to obtain the first motion track of the pedestrian includes: acquiring current state information and current motion control information of the pedestrian, wherein the current state information at least comprises position information and speed information, and the current motion control information at least comprises acceleration and angular velocity; establishing a kinematic model of the pedestrian based on the current state information and the current motion control information; and acquiring a first motion trail of the pedestrian through the kinematic model.
In an exemplary embodiment, the inputting the motion information of the pedestrian and other traffic participants into a second model to obtain a second motion track of the pedestrian comprises:
acquiring current position information of the pedestrians and other traffic participants; acquiring historical track information of the pedestrians and other traffic participants through an extraction model; acquiring interaction relation information of the pedestrian, other traffic participants and obstacles through an interaction model based on the current position information and the historical track information; and acquiring a second motion trail of the pedestrian based on the interactive relation information.
In an exemplary embodiment, before the acquiring the current scene where the pedestrian is located, the method further includes: and acquiring the state information of the obstacle.
In one exemplary embodiment, the state information of the obstacle includes at least historical position, type, and shape information of the obstacle
In a second aspect, an embodiment of the present disclosure further provides an apparatus for determining a motion trajectory of a pedestrian, including:
the first acquisition module is used for acquiring the current scene of the pedestrian; the second obtaining module is used for inputting the motion information of the pedestrian to a first model to obtain a first motion track of the pedestrian when the current scene is a first scene; and the third acquisition module is used for inputting the motion information of the pedestrian and other traffic participants into a second model to acquire a second motion track of the pedestrian when the current scene is a second scene.
In a third aspect, an embodiment of the present disclosure further provides a computer-readable storage medium, where the storage medium stores a computer program, and the computer program is configured to execute the determining method in any of the above technical solutions.
In a fourth aspect, an embodiment of the present disclosure further provides an electronic device, where the electronic device includes: a processor and a memory for storing processor-executable instructions; wherein the processor is configured to execute the determining method in any of the above technical solutions.
According to the above content, the pedestrian motion trajectory determination method provided by the embodiment of the disclosure combines the features of the high-precision map, so that the pedestrian motion prediction in the pedestrian crossing scene is simplified; meanwhile, interaction of pedestrians and other obstacles is considered in other scenes, historical motion features of the pedestrians are extracted by using a deep learning model, pedestrian trajectory learning modeling is carried out on a large-scale data set, the accuracy of a pedestrian prediction result is improved, and the robustness and the real-time performance of model prediction are improved on the premise of not losing the pedestrian prediction accuracy through the multi-mode pedestrian trajectory determination method.
In order to make the aforementioned objects, features and advantages of the present disclosure more comprehensible, preferred embodiments accompanied with figures are described in detail below.
Drawings
In order to more clearly illustrate the embodiments of the present disclosure or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly introduced below, it is obvious that the drawings in the following description are only some embodiments described in the present disclosure, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 is a flow chart of the steps of a method for predicting a pedestrian motion trajectory provided by the present disclosure;
FIG. 2 is a flowchart of the steps provided by the present disclosure to obtain the current scene in which the pedestrian is located;
FIG. 3 is a flowchart of the steps provided by the present disclosure to obtain a first motion profile of the pedestrian;
FIG. 4 is a flowchart of the steps provided by the present disclosure to obtain a second motion profile of the pedestrian;
FIG. 5 is a layout of a prediction of pedestrian trajectories through a neural network as provided by the present disclosure;
fig. 6 is a block diagram of a device for determining a pedestrian movement locus provided by the present disclosure;
fig. 7 is a block diagram of an electronic device provided by the present disclosure.
Detailed Description
Specific embodiments of the present disclosure will be described in detail below with reference to the accompanying drawings, but the present disclosure is not limited thereto.
It will be understood that various modifications may be made to the embodiments disclosed herein. Accordingly, the foregoing description should not be construed as limiting, but merely as exemplifications of embodiments. Other modifications will occur to those skilled in the art within the scope and spirit of the disclosure.
The accompanying drawings, which are incorporated in and constitute a part of the specification, illustrate embodiments of the disclosure and, together with a general description of the disclosure given above, and the detailed description of the embodiments given below, serve to explain the principles of the disclosure.
These and other characteristics of the present disclosure will become apparent from the following description of preferred forms of embodiment, given as non-limiting examples, with reference to the attached drawings.
It is also to be understood that although the present disclosure has been described with reference to specific examples, numerous other equivalent forms of practicing the disclosure as would be recognized by those skilled in the art are also within the scope of protection thereby defined.
The above and other aspects, features and advantages of the present disclosure will become more apparent in view of the following detailed description when taken in conjunction with the accompanying drawings.
The specification may use the phrases "in one embodiment," "in another embodiment," "in yet another embodiment," or "in other embodiments," which may each refer to one or more of the same or different embodiments in accordance with the disclosure.
The present disclosure is further described with reference to the following figures and specific examples.
Example 1
The embodiment of the disclosure relates to the technical field of computer vision, and in particular relates to a method for determining a pedestrian motion trajectory, wherein the method for determining the motion trajectory can be applied to an intelligent terminal, and the intelligent terminal is a movable intelligent terminal, such as: robots, intelligent wheelchairs, unmanned vehicles, and the like.
As shown in fig. 1, the prediction method flow described in the present disclosure includes steps S101 to S103. The specific realization principle of each step is as follows:
and S101, acquiring the current scene of the pedestrian.
In this step, a current scene where the pedestrian is located is obtained, where the scene where the pedestrian is located includes a scene where a target of the pedestrian is relatively clear, for example, the pedestrian is on a pedestrian crossing or a traffic sidewalk, and may also include the pedestrian and other traffic participants existing at the same time, and the target of the pedestrian is not clear enough and is in a weak place of the traffic participants, for example, the pedestrian is on a non-pedestrian crossing, on a rural road, and the like.
As shown in fig. 2, the acquiring of the current scene of the pedestrian provided by the present disclosure specifically includes the following steps:
s201, it is determined whether map information can be acquired at the current time.
In this step, it is determined whether map information, from which the position of the pedestrian can be determined, can be acquired at the current time.
And S202, judging whether the current position of the pedestrian is located in a preset area or not under the condition that the map information can be acquired.
After the above step S201 is completed, in this step, in the case where the map information can be acquired, it is determined whether the current position of the pedestrian is located within a predetermined area, where the predetermined area is a scene where the target of the pedestrian is relatively clear, for example, a pedestrian is on a sidewalk.
Specifically, the intelligent terminal may obtain map information from the server, and determine the current position of the pedestrian based on the map information, where the current position may be a scene in which the target of the pedestrian such as a sidewalk is relatively clear, or a scene in which the pedestrian such as a non-sidewalk and other traffic participants coexist.
S203, under the condition that the current position of the pedestrian is located in the preset area, the current scene is a first scene.
After the step S202 is completed, in this step, under the condition that the map information is acquired, under the condition that the current position of the pedestrian is located in a scene where the target of the pedestrian is relatively clear, such as a sidewalk, the current scene is a first scene; and under the condition that the current position of the pedestrian is located outside the preset area, the current scene is a second scene.
Specifically, in the second scenario, the pedestrian and other traffic participants exist simultaneously, the target of the pedestrian is not clear enough and is in the weak part of the traffic participants, for example, the pedestrian is on a non-sidewalk, on a rural road, etc., and at this time, the motion interaction between the pedestrian and other obstacles needs to be considered to avoid collision.
S102, when the current scene is a first scene, inputting the motion information of the pedestrian to a first model to obtain a first motion track of the pedestrian.
After the step S101 is completed, in this step, when the current scene is a first scene, the motion information of the pedestrian is input to a first model to obtain a first motion trajectory of the pedestrian.
Specifically, in consideration of the fact that the pedestrian has a relatively clear target in a scene such as a sidewalk, and the target of the pedestrian is clear in the scene, the direction based on the target guidance is adopted for research, and the unicycle model is adopted in the disclosure to predict the motion trail of the pedestrian.
As shown in fig. 3, the acquiring of the first motion trajectory of the pedestrian specifically includes the following steps:
s301, acquiring current state information and current motion control information of the pedestrian.
In this step, current state information and current motion control information of the pedestrian are obtained, where the current state information at least includes position information and velocity information, and the current motion control information at least includes acceleration and angular velocity.
Specifically, the state of the pedestrian is marked as X ∈ R 4 And the control information is marked as u e R 2 And simultaneously processing noise marked as w-N (0, W) epsilon R 4 The details are defined as follows:
where x, y are marked as spatial position coordinates of the pedestrian, the velocity is marked v, the direction of the velocity is marked theta, and the acceleration and angular velocity are marked a and omega, respectively, by which the physical state at a given moment can be represented.
S302, establishing a kinematics model of the pedestrian based on the S current state information and the current motion control information.
After the above step S301 is completed, in this step, a kinematic model of the pedestrian is established based on the current state information and the current motion control information.
Specifically, based on the current state information and the current motion control information of the pedestrian, a kinematics model of the pedestrian is established, which specifically includes:
and S303, acquiring a first motion track of the pedestrian through the kinematic model.
After obtaining the kinematic model of the pedestrian, in this step, a first motion trajectory of the pedestrian is obtained based on the kinematic model.
Specifically, the state information of the pedestrian at the next moment is calculated according to the position, the speed and the angle information of the pedestrian at the current moment; and further iterating to obtain all state information of the required prediction length. The modeling is carried out on the pedestrian by using the kinematics model, so that the motion prediction processing of the pedestrian in a pedestrian crossing scene is simplified, and the performance overhead of predicting the crowd by using a deep learning model is avoided.
S103, when the current scene is a second scene, inputting the motion information of the pedestrian and other traffic participants into a second model to obtain a second motion track of the pedestrian.
In this step, when the intelligent terminal judges that the current scene is a second scene, the motion information of the pedestrian is input to a second model to obtain a second motion track of the pedestrian.
Specifically, considering that when a pedestrian is in a non-pedestrian crosswalk scene and a high-precision map scene of an area which cannot be covered, the pedestrian is not clearly targeted and is in the weak side of a traffic participant, if the motion prediction is performed on the part of the pedestrian, the motion interaction between the pedestrian and other obstacles needs to be considered to avoid collision, so that at this time, the state information of the obstacle should be acquired in advance, wherein the state information of the obstacle at least includes the historical position, type and shape information of the obstacle.
As shown in fig. 4, the step of acquiring the second motion trajectory of the pedestrian specifically includes the following steps:
s401, acquiring the current position information of the pedestrian and other traffic participants.
In this step, the current position information of the pedestrian and other traffic participants is acquired.
S402, obtaining the historical track information of the pedestrians and other traffic participants through an extraction model.
In the step, historical track information of the pedestrians and other traffic participants is obtained through an extraction model. Specifically, as shown in fig. 5, the historical observation trajectory of the pedestrian is modeled by a deep learning model, wherein the output of the deep learning network is a time feature map, and the time feature map is composed of 3 groups of one-dimensional convolutions, each group has 2 residual blocks, and the step size of the first block is 2. And then fusing multi-scale features by using a Feature Pyramid Network (FPN), obtaining an output tensor by using another residual block, wherein the sizes of convolution kernels of all layers are 3, the number of output channels is 128, and layer normalization and ReLU operation are performed after each convolution operation. Extracting the historical track of the pedestrian through a CNN and FPN network, wherein the historical observation track of the pedestrian is characterized as { delta p -(T-1) ，...，Δp -1 ，Δp 0 In which Δ p is t Is two-dimensional coordinate information from time T-1 to time T, and T is the length of the observation trajectory, and is filled with 0 when the observation length is less than T. And simultaneously marking whether the track coordinate is filled with 1 × T, so that the size of the input tensor is 3 × T. The trace input is then processed using 1D CNN to improve its effectiveness in extracting multi-scale features and the efficiency of parallel computations.
And S403, acquiring interactive relationship information of the pedestrian, other traffic participants and obstacles through an interactive model based on the current position information and the historical track information.
After the steps S401 and S402 are completed, in this step, the interaction relationship information between the pedestrian and other traffic participants and obstacles is obtained through an interaction model based on the current position information and the historical track information.
Specifically, after the observation trajectory information features of each pedestrian and the traffic participant are extracted by the network, the interaction relationship of the traffic participants is processed by using another network. The network consists of two residual blocks, each containing an attention layer and a linear layer, and residual connections, and the output channels of each layer are also 128. The attention layer is used to characterize the interaction of traffic participants, and given a participant node i, the aggregate features with another participant j are defined as follows:
wherein x i For the characteristics of the i-node, W is the weight matrix, Ψ is the combination of the normalization layer and the ReLU operation, Δ i，j ＝MLP(v j -v i ) And v denotes coordinate information of the node.
S404, acquiring a second motion trail of the pedestrian based on the interaction relation information.
After the step S403 is completed, in this step, after the 128-dimensional output channel is obtained through the two networks, a prediction head is added to obtain a predicted trajectory.
The pedestrian motion trajectory determination method provided by the embodiment of the disclosure combines the characteristics of a high-precision map, and simplifies the pedestrian motion prediction in a pedestrian crossing scene; meanwhile, interaction of pedestrians and other obstacles is considered in other scenes, historical motion features of the pedestrians are extracted by using a deep learning model, pedestrian trajectory learning modeling is carried out on a large-scale data set, the accuracy of a pedestrian prediction result is improved, and the robustness and the real-time performance of model prediction are improved on the premise of not losing the pedestrian prediction accuracy through the multi-mode pedestrian trajectory determination method.
Example 2
To better implement the above method, the second aspect of the present disclosure also provides a determination apparatus of a pedestrian motion trajectory, which may be integrated on an electronic device.
For example, as shown in fig. 6, the device 200 for determining the pedestrian motion trajectory may include: the first obtaining module 210, the second obtaining module 220, and the third obtaining module 230 are as follows:
(1) A first obtaining module 210, configured to obtain a current scene where the pedestrian is located;
(2) A second obtaining module 220, configured to, when the current scene is a first scene, input motion information of the pedestrian to a first model to obtain a first motion trajectory of the pedestrian;
a third obtaining module 230, configured to, when the current scene is a second scene, input the motion information of the pedestrian and the other traffic participants into a second model to obtain a second motion trajectory of the pedestrian.
The pedestrian motion trajectory determination device provided by the embodiment of the disclosure combines the characteristics of a high-precision map, and simplifies the pedestrian motion prediction in a pedestrian crossing scene; meanwhile, interaction of pedestrians and other obstacles is considered in other scenes, historical motion features of the pedestrians are extracted by using a deep learning model, pedestrian trajectory learning modeling is carried out on a large-scale data set, the accuracy of a pedestrian prediction result is improved, and the robustness and the real-time performance of model prediction are improved on the premise of not losing the pedestrian prediction accuracy through the multi-mode pedestrian trajectory determination method.
Example 3
It will be understood by those skilled in the art that all or part of the steps of the methods of the above embodiments may be performed by instructions or by associated hardware controlled by the instructions, which may be stored in a computer readable storage medium and loaded and executed by a processor.
To this end, a third embodiment of the present disclosure provides a storage medium, which is a computer-readable medium storing a computer program, where the computer program is executed by a processor to implement the motion trajectory determination method provided in the embodiments of the present disclosure, and the method includes the following steps S11 to S13:
s11, acquiring a current scene where the pedestrian is located;
s12, when the current scene is a first scene, inputting the motion information of the pedestrian to a first model to obtain a first motion track of the pedestrian;
and S13, when the current scene is a second scene, inputting the motion information of the pedestrian and other traffic participants into a second model to obtain a second motion track of the pedestrian.
Further, the computer program, when executed by a processor, implements the other methods provided by any of the above embodiments of the present disclosure.
The pedestrian motion trajectory determination method provided by the embodiment of the disclosure combines the characteristics of a high-precision map, and simplifies the pedestrian motion prediction in a pedestrian crossing scene; meanwhile, interaction of pedestrians and other obstacles is considered in other scenes, historical motion features of the pedestrians are extracted by using a deep learning model, pedestrian trajectory learning modeling is carried out on a large-scale data set, the accuracy of a pedestrian prediction result is improved, and the robustness and the real-time performance of model prediction are improved on the premise of not losing the pedestrian prediction accuracy through the multi-mode pedestrian trajectory determination method.
Example 4
A fourth embodiment of the present disclosure provides an electronic device, as shown in fig. 7, the electronic device includes at least a processor 401 and a memory 402, the memory 402 stores a computer program thereon, and the processor 401, when executing the computer program on the memory 402, implements the pedestrian trajectory determination method provided in any embodiment of the present disclosure. Illustratively, the method performed by the electronic device computer program is as follows:
s21, acquiring a current scene where the pedestrian is located;
s22, when the current scene is a first scene, inputting the motion information of the pedestrian to a first model to obtain a first motion track of the pedestrian;
and S23, when the current scene is a second scene, inputting the motion information of the pedestrian and other traffic participants into a second model to obtain a second motion track of the pedestrian.
In a specific implementation, the first obtaining module 210, the second obtaining module 220, the determining module 230, and the like are stored in the memory 402 as program units, and the processor 401 executes the program units stored in the memory 402 to implement corresponding functions.
The pedestrian motion trajectory determination method provided by the embodiment of the disclosure combines the characteristics of a high-precision map, and simplifies the pedestrian motion prediction under the pedestrian crosswalk scene; meanwhile, interaction of pedestrians and other obstacles is considered in other scenes, historical motion features of the pedestrians are extracted by using a deep learning model, pedestrian trajectory learning modeling is carried out on a large-scale data set, the accuracy of a pedestrian prediction result is improved, and the robustness and the real-time performance of model prediction are improved on the premise of not losing the pedestrian prediction accuracy through the multi-mode pedestrian trajectory determination method.
The storage medium may be included in the electronic device; or may exist separately without being assembled into the electronic device.
The storage medium carries one or more programs that, when executed by the electronic device, cause the electronic device to: acquiring at least two internet protocol addresses; sending a node evaluation request comprising at least two internet protocol addresses to node evaluation equipment, wherein the node evaluation equipment selects the internet protocol addresses from the at least two internet protocol addresses and returns the internet protocol addresses; receiving an internet protocol address returned by the node evaluation equipment; wherein the obtained internet protocol address indicates an edge node in the content distribution network.
Alternatively, the storage medium carries one or more programs that, when executed by the electronic device, cause the electronic device to: receiving a node evaluation request comprising at least two internet protocol addresses; selecting an internet protocol address from at least two internet protocol addresses; returning the selected internet protocol address; wherein the received internet protocol address indicates an edge node in the content distribution network.
Computer program code for carrying out operations for the present disclosure may be written in any combination of one or more programming languages, including but not limited to an object oriented programming language such as Java, smalltalk, C + +, and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the passenger computer, partly on the passenger computer, as a stand-alone software package, partly on the passenger computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the passenger computer through any type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet service provider).
It should be noted that the storage media described above in this disclosure can be computer readable signal media or computer readable storage media or any combination of the two. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples of the computer readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the present disclosure, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. In contrast, in the present disclosure, a computer readable signal medium may comprise a propagated data signal with computer readable program code embodied therein, either in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may be any storage medium that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device. Program code embodied on a storage medium may be transmitted using any appropriate medium, including but not limited to: electrical wires, optical cables, RF (radio frequency), etc., or any suitable combination of the foregoing.
The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
The units described in the embodiments of the present disclosure may be implemented by software or hardware. Where the name of an element does not in some cases constitute a limitation on the element itself.
The functions described herein above may be performed, at least in part, by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field Programmable Gate Arrays (FPGAs), application Specific Integrated Circuits (ASICs), application Specific Standard Products (ASSPs), systems on a chip (SOCs), complex Programmable Logic Devices (CPLDs), and the like.
In the context of this disclosure, a machine-readable medium may be a tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
The foregoing description is only exemplary of the preferred embodiments of the disclosure and is illustrative of the principles of the technology employed. It will be appreciated by those skilled in the art that the scope of the disclosure herein is not limited to the particular combination of features described above, but also encompasses other embodiments in which any combination of the features described above or their equivalents does not depart from the spirit of the disclosure. For example, the above features and the technical features disclosed in the present disclosure (but not limited to) having similar functions are replaced with each other to form the technical solution.
Further, while operations are depicted in a particular order, this should not be understood as requiring that such operations be performed in the particular order shown or in sequential order. Under certain circumstances, multitasking and parallel processing may be advantageous. Likewise, while several specific implementation details are included in the above discussion, these should not be construed as limitations on the scope of the disclosure. Certain features that are described in the context of separate embodiments can also be implemented in combination in a single embodiment. Conversely, various features that are described in the context of a single embodiment can also be implemented in multiple embodiments separately or in any suitable subcombination.
While the present disclosure has been described in detail with reference to the embodiments, the present disclosure is not limited to the specific embodiments, and those skilled in the art can make various modifications and alterations based on the concept of the present disclosure, and the modifications and alterations should fall within the scope of the present disclosure as claimed.
Claims (10)
1. A method for determining a pedestrian motion trail is characterized by comprising the following steps:
acquiring a current scene where the pedestrian is located;
when the current scene is a first scene, inputting the motion information of the pedestrian to a first model to obtain a first motion track of the pedestrian;
and when the current scene is a second scene, inputting the motion information of the pedestrian and other traffic participants into a second model to obtain a second motion track of the pedestrian.
2. The determination method according to claim 1, wherein the acquiring the current scene where the pedestrian is located comprises:
judging whether the map information can be acquired at the current moment;
under the condition that map information can be acquired, judging whether the current position of the pedestrian is located in a preset area;
under the condition that the current position of the pedestrian is located in the predetermined area, the current scene is a first scene;
and under the condition that the current position of the pedestrian is located outside the preset area, the current scene is a second scene.
3. The determination method according to claim 2, further comprising:
and in the case of failing to acquire the map information, the current scene is a second scene.
4. The determination method according to claim 1, wherein the inputting the motion information of the pedestrian to a first model to obtain a first motion trajectory of the pedestrian comprises:
acquiring current state information and current motion control information of the pedestrian, wherein the current state information at least comprises position information and speed information, and the current motion control information at least comprises acceleration and angular velocity;
establishing a kinematic model of the pedestrian based on the current state information and the current motion control information;
and acquiring a first motion trail of the pedestrian through the kinematic model.
5. The determination method according to claim 1, wherein the inputting the motion information of the pedestrian and other traffic participants into a second model to obtain a second motion track of the pedestrian comprises:
acquiring current position information of the pedestrians and other traffic participants;
acquiring historical track information of the pedestrians and other traffic participants through an extraction model;
acquiring interaction relation information of the pedestrian, other traffic participants and obstacles through an interaction model based on the current position information and the historical track information;
and acquiring a second motion trail of the pedestrian based on the interaction relation information.
6. The determination method according to claim 5, further comprising, before the obtaining the current scene where the pedestrian is located: and acquiring the state information of the obstacle.
7. The determination method according to claim 5, characterized in that the state information of the obstacle includes at least historical position, type, and shape information of the obstacle.
8. A device for determining a trajectory of a pedestrian, comprising:
the first acquisition module is used for acquiring the current scene of the pedestrian;
the second obtaining module is used for inputting the motion information of the pedestrian to a first model to obtain a first motion track of the pedestrian when the current scene is a first scene;
and the third acquisition module is used for inputting the motion information of the pedestrian and other traffic participants into a second model to acquire a second motion track of the pedestrian when the current scene is a second scene.
9. A computer-readable storage medium, which stores a computer program for executing the determination method of any one of the above claims 1 to 7.
10. An electronic device, the electronic device comprising: a processor and a memory for storing processor-executable instructions; wherein the processor is configured to perform the determination method of any of the preceding claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211502611.7A CN115817531A (en) | 2022-11-25 | 2022-11-25 | Pedestrian motion track determination method and determination device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211502611.7A CN115817531A (en) | 2022-11-25 | 2022-11-25 | Pedestrian motion track determination method and determination device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115817531A true CN115817531A (en) | 2023-03-21 |
Family
ID=85532230
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211502611.7A Pending CN115817531A (en) | 2022-11-25 | 2022-11-25 | Pedestrian motion track determination method and determination device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115817531A (en) |
-
2022
- 2022-11-25 CN CN202211502611.7A patent/CN115817531A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Deo et al. | Multi-modal trajectory prediction of surrounding vehicles with maneuver based lstms | |
| CN111626208B (en) | Method and device for detecting small objects | |
| CN113272830B (en) | Trajectory representation in behavior prediction system | |
| EP3651064B1 (en) | Deep learning for object detection using pillars | |
| EP3822852B1 (en) | Method, apparatus, computer storage medium and program for training a trajectory planning model | |
| CN110356412B (en) | Method and apparatus for automatic rule learning for autonomous driving | |
| CN111316286A (en) | Trajectory prediction method and device, storage medium, driving system and vehicle | |
| WO2021040910A1 (en) | Occupancy prediction neural networks | |
| US11449705B2 (en) | Field theory based perception for autonomous vehicles | |
| CN111292352B (en) | Multi-target tracking method, device, equipment and storage medium | |
| CN112212874A (en) | Vehicle track prediction method and device, electronic equipment and computer readable medium | |
| CN114212110B (en) | Obstacle trajectory prediction method and device, electronic equipment and storage medium | |
| US11367289B1 (en) | Machine learning-based framework for drivable surface annotation | |
| WO2021017504A1 (en) | Method and apparatus for predicting motion trajectory | |
| WO2023201964A1 (en) | Vehicle following target determination method and apparatus, and device and medium | |
| CN113654564A (en) | Method for a vehicle | |
| CN115339453B (en) | Vehicle lane change decision information generation method, device, equipment and computer medium | |
| GB2608201A (en) | Identifying objects using LiDAR | |
| Houben et al. | Park marking-based vehicle self-localization with a fisheye topview system | |
| US11875680B2 (en) | Systems and methods for augmenting perception data with supplemental information | |
| CN114882457A (en) | Model training method, lane line detection method and equipment | |
| CN113139696B (en) | Trajectory prediction model construction method and trajectory prediction method and device | |
| CN115817531A (en) | Pedestrian motion track determination method and determination device | |
| US11946749B2 (en) | Driving data guided spatial planning | |
| CN115454082A (en) | Vehicle obstacle avoidance method and system, computer readable storage medium and electronic device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |