Internet of vehicles resource scheduling method and device for trajectory prediction, equipment and medium

Abstract

The application relates to a vehicle networking resource scheduling method and device for trajectory prediction, equipment and medium, wherein the method comprises the following steps: receiving current state information uploaded by each user at each time; extracting the time sequence characteristics of the state information, and performing trajectory accuracy estimation processing on each user at the current time and the next time based on the time sequence characteristics by using a target deep neural network module to obtain the current time trajectory accuracy and the next time trajectory accuracy; generating a user communication priority based on the current time trajectory accuracy and the next time trajectory accuracy, and performing vehicle networking resource scheduling for trajectory prediction according to the communication priority. By considering the performance of a single communication link while focusing on the maximization of overall trajectory prediction accuracy, the application alleviates the influence caused by the incomplete historical state information, solves the problem of reduced trajectory prediction accuracy under the condition of incomplete historical information, and improves the accuracy of trajectory prediction.

Description

Technical Field

[0001] This application relates to the field of vehicle-to-everything (V2X) communication technology, and in particular to a V2X resource scheduling method, apparatus, device, and medium for trajectory prediction. Background Technology

[0002] Vehicle trajectory prediction is a key technology in intelligent transportation systems, aiming to predict the position and route of vehicles over a future period using various algorithms. The core of this technology lies in a deep understanding of vehicle dynamics and behavioral patterns, utilizing historical data and real-time information to comprehensively determine the future motion state of vehicles. Vehicle trajectory prediction plays a crucial role in improving road safety, optimizing traffic flow management, reducing traffic congestion, and promoting the development of autonomous driving technology. Through efficient and accurate trajectory prediction, road safety can be significantly improved, traffic flow management optimized, congestion reduced, and the decision-making capabilities of autonomous vehicles enhanced.

[0003] In trajectory prediction technology, based on the task execution method, it can be divided into two scenarios: distributed and centralized. We focus specifically on centralized trajectory prediction tasks. These tasks are typically implemented in urban traffic management centers or large-scale intelligent traffic monitoring systems. Their core lies in using centralized processing to comprehensively analyze data from a wide range of vehicle networks in real time, forming a macro-level understanding and prediction of the entire traffic system. In this scenario, centralized trajectory prediction relies on vehicle-to-everything (V2X) communication technology. The central node acquires key status information of each vehicle within its coverage area in real time through V2X, such as location, speed, and other sensor data. This data provides the necessary input for trajectory prediction, enabling the prediction model to perform accurate real-time analysis based on comprehensive information.

[0004] However, due to the complex communication environment faced by vehicle-to-everything (V2X) communication, including the high dynamism of network topology caused by vehicle movement, spatiotemporal complexity, and the complex and rapidly changing wireless propagation environment, the reliable transmission of vehicle status information becomes a communication challenge. This challenge may lead to incomplete vehicle status information received by the vehicle trajectory prediction system, thereby affecting the central node's complete capture of historical vehicle status information, ultimately impacting the accuracy of prediction results, as well as the operational efficiency and safety of the entire intelligent transportation system. Summary of the Invention

[0005] The purpose of this application is to propose a method, device, equipment, and medium for vehicle network resource scheduling for trajectory prediction. By managing and scheduling limited communication resources, it achieves the best balance between successful communication transmission and trajectory prediction accuracy, improves the overall accuracy of the vehicle trajectory prediction system, and solves the problem of decreased trajectory prediction accuracy caused by incomplete historical state information due to unreliable transmission of vehicle network communication.

[0006] To address the aforementioned technical problems, this application provides a vehicle-to-everything (V2X) resource scheduling method for trajectory prediction, applied in a V2X wireless communication system, comprising:

[0007] Receive the current status information uploaded by each user at each moment;

[0008] The temporal features of the state information are extracted, and the target deep neural network module is used to perform trajectory accuracy estimation for each user at the current time and the next time based on the temporal features, so as to obtain the trajectory accuracy at the current time and the trajectory accuracy at the next time.

[0009] User communication priorities are generated based on the current trajectory accuracy and the next trajectory accuracy, and vehicle network communication resources are scheduled according to the user communication priorities.

[0010] To address the aforementioned technical problems, this application provides a vehicle-to-everything (V2X) resource scheduling device for trajectory prediction, applied in a V2X wireless communication system, comprising:

[0011] The status information receiving unit is used to receive the current status information of each user at each moment;

[0012] The trajectory accuracy generation unit is used to extract the temporal features of the state information, and use the target deep neural network module to perform trajectory accuracy estimation processing for each user at the current time and the next time based on the temporal features, so as to obtain the trajectory accuracy at the current time and the trajectory accuracy at the next time.

[0013] The communication resource scheduling unit is used to generate user communication priorities based on the trajectory accuracy at the current time and the trajectory accuracy at the next time, and to perform vehicle network communication resource scheduling according to the user communication priorities.

[0014] To solve the above-mentioned technical problems, one technical solution adopted by the present invention is to provide a computer device, including one or more processors; and a memory for storing one or more programs, so that the one or more processors implement the vehicle network resource scheduling method for trajectory prediction as described in any one of the above-mentioned methods.

[0015] To solve the above-mentioned technical problems, one technical solution adopted by the present invention is: a computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor, implements the vehicle network resource scheduling method for trajectory prediction described in any one of the above-mentioned methods.

[0016] This invention provides a method, apparatus, device, and medium for vehicle-to-everything (V2X) resource scheduling for trajectory prediction. The method includes: receiving current state information uploaded by each user at each moment; extracting temporal features from the state information; and using a target deep neural network module to estimate the trajectory accuracy of each user at the current moment and the next moment based on the temporal features, obtaining the trajectory accuracy at the current moment and the trajectory accuracy at the next moment; generating user communication priorities based on the current moment trajectory accuracy and the next moment trajectory accuracy; and scheduling V2X communication resources according to the user communication priorities. This invention estimates trajectory accuracy based on user-uploaded state information and calculates user communication priorities and schedules V2X communication resources based on the estimated trajectory accuracy values. It considers the performance of individual communication links while focusing on maximizing overall trajectory prediction accuracy, mitigating the impact of incomplete historical state information, solving the problem of reduced trajectory prediction accuracy under incomplete historical information, and improving the accuracy of trajectory prediction in intelligent transportation systems. Attached Figure Description

[0017] To more clearly illustrate the solutions in this application, the accompanying drawings used in the description of the embodiments of this application will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart illustrating the implementation of the vehicle network resource scheduling method for trajectory prediction provided in the embodiments of this application.

[0019] Figure 2 This is a flowchart illustrating the implementation of a sub-process in the vehicle network resource scheduling method for trajectory prediction provided in the embodiments of this application.

[0020] Figure 3 This is a flowchart illustrating the implementation of a sub-process in the vehicle network resource scheduling method for trajectory prediction provided in the embodiments of this application.

[0021] Figure 4 This is a schematic diagram of the trajectory accuracy estimation model provided in the embodiments of this application;

[0022] Figure 5 This is a flowchart illustrating the implementation of a sub-process in the vehicle network resource scheduling method for trajectory prediction provided in the embodiments of this application.

[0023] Figure 6 This is a flowchart illustrating the implementation of a sub-process in the vehicle network resource scheduling method for trajectory prediction provided in the embodiments of this application.

[0024] Figure 7This is a schematic diagram illustrating the average accuracy over time provided in the embodiments of this application;

[0025] Figure 8 This is another schematic diagram illustrating the performance of average accuracy over time provided in the embodiments of this application;

[0026] Figure 9 This is another schematic diagram illustrating the performance of average accuracy over time provided in the embodiments of this application;

[0027] Figure 10 This is another schematic diagram illustrating the performance of average accuracy over time provided in the embodiments of this application;

[0028] Figure 11 This is another schematic diagram illustrating the performance of average accuracy over time provided in the embodiments of this application;

[0029] Figure 12 This is a schematic diagram of a vehicle network resource scheduling device for trajectory prediction provided in an embodiment of this application;

[0030] Figure 13 This is a schematic diagram of the computer device provided in the embodiments of this application. Detailed Implementation

[0031] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains; the terminology used herein in the specification of the application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application; the terms "comprising" and "having," and any variations thereof, in the specification, claims, and foregoing drawings of this application, are intended to cover non-exclusive inclusion. The terms "first," "second," etc., in the specification, claims, or foregoing drawings of this application are used to distinguish different objects, not to describe a particular order.

[0032] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0033] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings.

[0034] The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0035] It should be noted that the vehicle network resource scheduling method for trajectory prediction provided in this application embodiment is applied to the vehicle network wireless communication system, and correspondingly, the vehicle network resource scheduling device for trajectory prediction is also applied to the vehicle network wireless communication system.

[0036] Please see Figure 1 , Figure 1 A specific implementation of a vehicle-to-everything (V2X) resource scheduling method for trajectory prediction is shown.

[0037] This application aims to manage and schedule limited communication resources in the complex and ever-changing vehicle-to-everything (V2X) communication environment to achieve an optimal balance between successful communication transmission and trajectory prediction accuracy, thereby significantly improving the overall accuracy of the vehicle trajectory prediction system. This application first uses an accuracy estimation module to comprehensively evaluate the specific impact of information incompleteness caused by unreliable communication transmission on trajectory prediction accuracy. Then, based on the accuracy estimation of each vehicle and the status of the communication link, the communication priority of the corresponding vehicle is adjusted. The priority evaluation mainly considers the probability of successful communication and the expected accuracy improvement after successful communication. The core of this application is to achieve a dynamic balance, considering not only the performance of individual communication links but also focusing on maximizing the overall trajectory prediction accuracy, thus solving the problem of decreased trajectory prediction accuracy caused by incomplete historical state information due to unreliable V2X communication transmission.

[0038] It should be noted that if substantially the same result is obtained, the method of this invention is not based on... Figure 1 Limited to the order of the processes shown, this method includes the following steps:

[0039] S1: Receive the current status information uploaded by each user at each moment.

[0040] Furthermore, the state information includes at least one of position, velocity, acceleration, and orientation angle.

[0041] In this embodiment of the application, the current status information s uploaded by each user at each time moment is received. n,t＝i ={x n,t＝i ,y n,t＝i vx n,t＝i ,vy n,t＝i ,ax n,t＝i ,ay n,t＝i ,θ n,t＝i}, where n represents the number of users, i represents the current timestamp, and x n,t＝i ,y n,t＝i For the current position, vx n,t＝i ,vy n,t＝i Let ax be the velocity at the current moment. n,t＝i ,ay n,t＝i Let θ be the acceleration at the current moment.n,t＝i This represents the heading angle at the current moment.

[0042] Please see Figure 2 , Figure 2 A specific implementation method following step S1 is shown below:

[0043] S1A: Store the status information uploaded by each user in a storage queue of a preset length corresponding to that user.

[0044] S1B: If the currently stored state information exceeds the window of the storage queue, delete the currently stored state information.

[0045] Specifically, the central node provides each user with a storage queue of preset length, storing the user's uploaded status information in timestamp format. With a fixed window length, the queue moves forward at each interval; if the currently stored status information exceeds the window size, it is deleted. It should be noted that the preset length is set according to actual conditions and is not limited here. In one specific embodiment, the central node provides each user with a fixed-length storage queue for storing the user's uploaded status information.

[0046] S2: Extract the temporal features of the state information, and use the target deep neural network module to perform trajectory accuracy estimation processing for each user at the current time and the next time based on the temporal features, so as to obtain the trajectory accuracy at the current time and the trajectory accuracy at the next time.

[0047] Specifically, in this embodiment of the application, the state information of each user in the storage queue is sent into the trajectory accuracy estimation model to estimate the trajectory accuracy of each user at the current time and the next time, thereby obtaining the trajectory accuracy at the current time and the trajectory accuracy at the next time.

[0048] Please see Figure 3 and Figure 4 , Figure 3 One specific implementation of step S2 is shown; Figure 4 This is a schematic diagram of the trajectory accuracy estimation model provided in the embodiments of this application; details are as follows:

[0049] S21: The state information is passed through multiple gate loop units to capture the timing characteristics of the state information.

[0050] Specifically, the state information passes through multiple Gate Recurrent Units (GRUs) to capture the temporal characteristics and long-term dependencies of the state information, transmit and utilize information in the sequence, and prevent gradient vanishing or gradient exploding problems.

[0051] The gated recurrent unit (GRU) is a variant of the recurrent neural network (RNN) specifically designed to process sequential data and help capture long-term dependencies within the data. The GRU maintains and updates its internal state by controlling the flow of information, thus addressing the vanishing and exploding gradient problems common in traditional RNNs. The GRU comprises the following key steps:

[0052] Reset Gate: Controls which old information should be forgotten. It determines how historical information should be considered in the current time step.

[0053] Update Gate: Determines the extent to which new candidate values ​​are added to the candidate value set. It determines how much historical information should be retained in the current time step.

[0054] Candidate Value: Based on the input data and the new information obtained from the reset gate, calculate the candidate value that can be added to the current state.

[0055] Furthermore, in trajectory prediction scenarios, gated loop units (GLUs) help the model remember and utilize past state information, such as the vehicle's historical trajectory, speed, and acceleration, thereby better predicting the vehicle's future trajectory. By adjusting and controlling the reset and update gates, GLUs can more effectively handle long-term dependencies, improving the model's performance and accuracy on sequential data.

[0056] Current State: The current state calculated based on the update gate and candidate values. This state will be passed to the next time step or the next layer of the network.

[0057] S22: Input the time-series features into the target deep neural network module for feature extraction to obtain nonlinear features.

[0058] Specifically, the extracted temporal features are input into the target deep neural network module for learning to capture complex nonlinear relationships between data, perform more refined feature representation and pattern learning on the data, and thus obtain nonlinear features.

[0059] Please see Figure 5 , Figure 5 A specific implementation of step S22 is shown below:

[0060] S221: Input the temporal features into the target deep neural network module.

[0061] S222: The temporal features are weighted and combined through each hidden layer in the target deep neural network module, and the weighted and combined temporal features are nonlinearly processed by an activation function to obtain the nonlinear features.

[0062] Specifically, the deep neural network module includes multiple hidden layers, and in each hidden layer, the temporal features of the input are weighted and combined, and nonlinearity is introduced through activation functions to generate a higher level of abstract feature representation, thereby obtaining nonlinear features.

[0063] Furthermore, before implementing step S2, this embodiment of the application also trains the deep neural network module to obtain the target deep neural network module. The model training includes: (1) Feature extraction: Input data is preprocessed and then fed into the DNN module. This data consists of feature representations or other relevant data processed in step 21. (2) Multiple hidden layers: The DNN contains multiple hidden layers, each composed of multiple neurons. Each hidden layer weights and combines the input data, and introduces nonlinearity through an activation function to generate higher-level abstract feature representations. (3) Backpropagation and optimization: The DNN is trained using a backpropagation algorithm. During training, the network parameters are continuously adjusted using optimization algorithms (such as gradient descent) to gradually reduce the error between the model's prediction and the actual result. Loss function calculation: The difference between the model's prediction and the actual result is evaluated using a loss function. During backpropagation, the optimizer uses these differences to update the weights and biases of the neural network.

[0064] S23: Perform residual connection between the state information and the nonlinear feature to output the current trajectory accuracy and the next trajectory accuracy for each user.

[0065] Specifically, residual connections between state information and nonlinear features help information propagate better in deep networks, reduce information loss, alleviate the vanishing gradient problem, and accelerate the training process. Further residual connections between state information and nonlinear features output the current and next time-step trajectory accuracy for each user.

[0066] S3: Generate user communication priorities based on the current trajectory accuracy and the next trajectory accuracy, and schedule vehicle network communication resources according to the user communication priorities.

[0067] The trajectory accuracy at the next moment includes the trajectory accuracy at the next moment after successful transmission and the trajectory accuracy at the next moment after transmission failure.

[0068] Specifically, in the implementation of this application, it is necessary to obtain the current channel conditions, then calculate the benefit of scheduling vehicles at the current time based on the current trajectory accuracy and the next trajectory accuracy of each user, then prioritize user communication based on the benefit of scheduling vehicles at the current time, and perform vehicle network communication resource scheduling based on the user communication priority.

[0069] Please see Figure 6 , Figure 6 A specific implementation of step S3 is shown below:

[0070] S31: Calculate the communication transmission success rate at the current moment.

[0071] S32: Calculate the revenue of dispatching vehicles at the current moment based on the communication transmission success rate, the trajectory accuracy at the current moment, the trajectory accuracy at the next moment after successful transmission, and the trajectory accuracy at the next moment after failed transmission, according to the first preset formula.

[0072] S33: Calculate the user communication priority based on the revenue according to the second preset formula, and schedule vehicle network communication resources according to the user communication priority.

[0073] In this embodiment, the central node can schedule resources to select one user for communication at any given time. Considering channel characteristics such as multipath effects in the vehicle network, the channel is modeled as a Rayleigh fading model, and the probability density function of the received signal amplitude is:

[0074]

[0075] Where σ is the fading parameter, used to describe the signal strength. When the signal strength x ≥ γ, the communication transmission is considered successful; otherwise, it is considered a transmission failure. γ is the reception threshold. It should be noted that the reception threshold is set according to the actual situation and is not limited here.

[0076] Specifically, in this embodiment, the communication transmission success rate is calculated based on the fading parameters at the current moment. Then, the revenue of dispatching vehicles at the current moment is calculated based on the communication transmission success rate, the trajectory accuracy at the current moment, the trajectory accuracy at the next moment after successful transmission, and the trajectory accuracy at the next moment after failed transmission, according to the first preset formula. The user communication priority is calculated based on the revenue according to the second preset formula, and the vehicle network communication resources are scheduled according to the user communication priority.

[0077] Furthermore, the first preset formula and the second preset formula are respectively:

[0078]

[0079]

[0080] Where, ε n,0 The trajectory accuracy at the current moment is... The trajectory accuracy at the next moment after successful transmission. Let p be the trajectory accuracy at the next moment after the transmission failure, and p be the communication transmission success rate. For the stated benefit, a t Select the user to be scheduled first at time t.

[0081] like Figures 7 to 11 As shown, this figure compares the changes in the global average error with the scheduling error at each time step for different scenarios, different vehicles, and different scheduling methods. Specifically, simulations validate the use of different methods for communication resource scheduling in different scenarios on the Nuscenes dataset, where baseline method 1 considers the current accuracy ε. n,0 The worst-case scenario is vehicle-priority communication. Benchmark method 2 considers prioritizing communication with the vehicle that has the highest current communication success rate. Benchmark method 3 considers randomly selecting one of the channels for communication. The changing trends of system trajectory prediction accuracy under different strategies in continuous time-series scheduling tasks are evaluated. Figures 7 to 11 It can be seen that the scheduling error of the method proposed in this embodiment is significantly lower than that of the other three benchmark methods, which verifies the effectiveness of the method proposed in this embodiment. Furthermore, it can be observed that the global average error of the method proposed in this embodiment decreases over time, which verifies the effectiveness of the scheduling method proposed in this embodiment.

[0082] In this embodiment, the system receives the current status information uploaded by each user at each moment; extracts the temporal features of the status information, and uses a target deep neural network module to estimate the trajectory accuracy of each user at the current moment and the next moment based on the temporal features, obtaining the trajectory accuracy at the current moment and the trajectory accuracy at the next moment; generates user communication priorities based on the trajectory accuracy at the current moment and the trajectory accuracy at the next moment, and schedules vehicle network communication resources according to the user communication priorities. This embodiment of the invention estimates trajectory accuracy based on the status information uploaded by users, and calculates communication priorities and schedules vehicle network communication resources based on the estimated trajectory accuracy values. By considering the performance of individual communication links while focusing on maximizing the overall trajectory prediction accuracy, it mitigates the impact of incomplete historical status information, thereby optimizing the accuracy of trajectory prediction in the entire system. Please refer to... Figure 12 As a response to the above Figure 1 The implementation of the method shown in this application provides an embodiment of a vehicle-to-everything (V2X) resource scheduling device for trajectory prediction. This device embodiment is similar to... Figure 1 Corresponding to the method embodiments shown, this device can be specifically applied to various electronic devices.

[0083] like Figure 12 As shown, the vehicle network resource scheduling device for trajectory prediction in this embodiment includes: a status information receiving unit 41, a trajectory accuracy generation unit 42, and a communication resource scheduling unit 43, wherein:

[0084] The status information receiving unit 41 is used to receive the current status information of each user at each moment;

[0085] The trajectory accuracy generation unit 42 is used to extract the temporal features of the state information, and use the target deep neural network module to perform trajectory accuracy estimation processing for each user at the current time and the next time based on the temporal features, so as to obtain the trajectory accuracy at the current time and the trajectory accuracy at the next time.

[0086] The communication resource scheduling unit 43 is used to generate user communication priorities based on the current trajectory accuracy and the next trajectory accuracy, and to perform vehicle network communication resource scheduling according to the user communication priorities.

[0087] Furthermore, the trajectory accuracy generation unit 42 includes:

[0088] A timing feature capture unit is used to pass the state information through multiple gate loop units to capture the timing features of the state information;

[0089] A nonlinear feature generation unit is used to input the temporal features into the target deep neural network module for feature extraction to obtain nonlinear features;

[0090] The residual connection unit is used to perform residual connection between the state information and the nonlinear feature to output the current trajectory accuracy and the next trajectory accuracy for each user.

[0091] Furthermore, the nonlinear feature generation unit includes:

[0092] A temporal feature input unit is used to input the temporal features into the target deep neural network module;

[0093] The nonlinear processing unit is used to perform weighted combination processing on the temporal features through each hidden layer in the target deep neural network module, and to perform nonlinear processing on the weighted combination temporal features through an activation function to obtain the nonlinear features.

[0094] Furthermore, the communication resource scheduling unit 43 includes:

[0095] The communication transmission success rate calculation unit is used to calculate the communication transmission success rate at the current moment.

[0096] The revenue calculation unit is used to calculate the revenue of the dispatched vehicle at the current moment based on the communication transmission success rate, the trajectory accuracy at the current moment, the trajectory accuracy at the next moment after successful transmission, and the trajectory accuracy at the next moment after failed transmission, according to a first preset formula.

[0097] The user communication priority calculation unit is used to calculate the user communication priority based on the revenue according to the second preset formula, and to schedule vehicle network communication resources according to the user communication priority.

[0098] Furthermore, the first preset formula and the second preset formula are respectively:

[0099]

[0100]

[0101] Where, ε n,0 The trajectory accuracy at the current moment is... The trajectory accuracy at the next moment after successful transmission. Let p be the trajectory accuracy at the next moment after the transmission failure, and p be the communication transmission success rate. For the stated benefit, a t Select the user to be scheduled first at time t.

[0102] Furthermore, the status information receiving unit 41 also includes:

[0103] A status information storage unit is used to store the status information uploaded by each user in a storage queue of a preset length corresponding to that user.

[0104] The status information deletion unit is used to delete the currently stored status information if the currently stored status information exceeds the window of the storage queue.

[0105] Furthermore, the state information includes at least one of position, velocity, acceleration, and orientation angle.

[0106] In this embodiment, the system receives the current state information uploaded by each user at each moment; extracts the temporal features of the state information, and uses a target deep neural network module to estimate the trajectory accuracy of each user at the current moment and the next moment based on the temporal features, obtaining the trajectory accuracy at the current moment and the trajectory accuracy at the next moment; generates user communication priorities based on the trajectory accuracy at the current moment and the trajectory accuracy at the next moment, and schedules vehicle network communication resources according to the user communication priorities. This embodiment of the invention estimates trajectory accuracy based on the state information uploaded by users, and calculates user communication priorities and schedules vehicle network communication resources based on the estimated trajectory accuracy values. It considers the performance of individual communication links while focusing on maximizing the overall trajectory prediction accuracy, mitigating the impact of incomplete historical state information, solving the problem of reduced trajectory prediction accuracy under incomplete historical information, and improving the accuracy of trajectory prediction in intelligent transportation systems.

[0107] To address the aforementioned technical problems, embodiments of this application also provide a computer device. Please refer to [link / reference needed]. Figure 13 , Figure 13 This is a basic structural block diagram of the computer device in this embodiment.

[0108] Computer device 5 includes a memory 51, a processor 52, and a network interface 53 that are interconnected via a system bus. It should be noted that only a computer device 5 with three components—memory 51, processor 52, and network interface 53—is shown in the figure; however, it should be understood that it is not required to implement all the components shown, and more or fewer components can be implemented alternatively. Those skilled in the art will understand that the computer device described here is a device capable of automatically performing numerical calculations and / or information processing according to pre-set or stored instructions, and its hardware includes, but is not limited to, microprocessors, application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), digital signal processors (DSPs), embedded devices, etc.

[0109] Computer devices can include desktop computers, laptops, handheld computers, and cloud servers. These devices allow for human-computer interaction with users through keyboards, mice, remote controls, touchpads, or voice-activated devices.

[0110] The memory 51 includes at least one type of readable storage medium, including flash memory, hard disk, multimedia card, card-type memory (e.g., SD or DX memory), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, magnetic disk, optical disk, etc. In some embodiments, the memory 51 may be an internal storage unit of the computer device 5, such as the hard disk or memory of the computer device 5. In other embodiments, the memory 51 may also be an external storage device of the computer device 5, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc., equipped on the computer device 5. Of course, the memory 51 may include both the internal storage unit and the external storage device of the computer device 5. In this embodiment, the memory 51 is typically used to store the operating system and various application software installed on the computer device 5, such as program code for a vehicle network resource scheduling method for trajectory prediction. In addition, the memory 51 can also be used to temporarily store various types of data that have been output or will be output.

[0111] In some embodiments, processor 52 may be a central processing unit (CPU), controller, microcontroller, microprocessor, or other data processing chip. This processor 52 is typically used to control the overall operation of computer device 5. In this embodiment, processor 52 is used to run program code stored in memory 51 or process data, for example, to run the program code of the above-described vehicle network resource scheduling method for trajectory prediction, to implement various embodiments of the vehicle network resource scheduling method for trajectory prediction.

[0112] The network interface 53 may include a wireless network interface or a wired network interface, which is typically used to establish a communication connection between the computer device 5 and other electronic devices.

[0113] This application also provides another embodiment, namely, providing a computer-readable storage medium storing a computer program that can be executed by at least one processor to cause the at least one processor to perform the steps of the above-described vehicle network resource scheduling method for trajectory prediction.

[0114] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods of the various embodiments of this application.

[0115] Obviously, the embodiments described above are only some embodiments of this application, not all embodiments. The accompanying drawings show preferred embodiments of this application, but do not limit the patent scope of this application. This application can be implemented in many different forms; rather, the purpose of providing these embodiments is to provide a more thorough and comprehensive understanding of the disclosure of this application. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing specific embodiments, or make equivalent substitutions for some of the technical features. Any equivalent structures made using the content of this application's specification and drawings, directly or indirectly applied to other related technical fields, are similarly within the scope of patent protection of this application.

Claims

1. A vehicle-to-everything (V2X) resource scheduling method for trajectory prediction, applied in a V2X wireless communication system, characterized in that, include: Receive the current status information uploaded by each user at each moment; The temporal features of the state information are extracted, and the target deep neural network module is used to perform trajectory accuracy estimation for each user at the current time and the next time based on the temporal features, so as to obtain the trajectory accuracy at the current time and the trajectory accuracy at the next time. User communication priorities are generated based on the current trajectory accuracy and the next trajectory accuracy, and vehicle network communication resources are scheduled according to the user communication priorities. The next-moment trajectory accuracy includes the trajectory accuracy after successful transmission and the trajectory accuracy after transmission failure. The step of generating user communication priorities based on the current-moment trajectory accuracy and the next-moment trajectory accuracy, and scheduling vehicle-to-everything (V2X) communication resources according to the user communication priorities, includes: Calculate the communication transmission success rate at the current moment; The revenue of dispatching vehicles at the current moment is calculated based on the communication transmission success rate, the trajectory accuracy at the current moment, the trajectory accuracy at the next moment after successful transmission, and the trajectory accuracy at the next moment after failed transmission, according to the first preset formula. The user communication priority is calculated based on the revenue according to the second preset formula, and vehicle network communication resources are scheduled according to the user communication priority. The first preset formula and the second preset formula are as follows: ； ； in, The trajectory accuracy at the current moment is... The trajectory accuracy at the next moment after successful transmission. The trajectory accuracy at the next moment after the transmission failure. The communication transmission success rate, For the aforementioned benefits, for Users who are selected for priority scheduling at all times.

2. The vehicle-to-everything (V2X) resource scheduling method for trajectory prediction according to claim 1, characterized in that, The extraction of temporal features from the state information, and the use of a target deep neural network module to estimate the trajectory accuracy of each user at the current and next time moments based on the temporal features, to obtain the trajectory accuracy at the current and next time moments, includes: The state information is passed through multiple gate loop units to capture the temporal characteristics of the state information; The temporal features are input into the target deep neural network module for feature extraction to obtain nonlinear features; The state information and the nonlinear feature are residually concatenated to output the current trajectory accuracy and the next trajectory accuracy for each user.

3. The vehicle network resource scheduling method for trajectory prediction according to claim 2, characterized in that, The step of inputting the temporal features into the target deep neural network module for feature extraction to obtain nonlinear features includes: The temporal features are input into the target deep neural network module; The temporal features are weighted and combined by each hidden layer in the target deep neural network module, and then nonlinearly processed by an activation function to obtain the nonlinear features.

4. The vehicle-to-everything (V2X) resource scheduling method for trajectory prediction according to any one of claims 1 to 3, characterized in that, After receiving the current status information uploaded by each user at each time moment, the method further includes: The status information uploaded by each user is stored in a storage queue of a preset length corresponding to that user; If the currently stored state information exceeds the window of the storage queue, the currently stored state information will be deleted.

5. The vehicle-to-everything (V2X) resource scheduling method for trajectory prediction according to any one of claims 1 to 3, characterized in that, The status information includes at least one of position, velocity, acceleration, and orientation angle.

6. A vehicle-to-everything (V2X) resource scheduling device for trajectory prediction, applied in a V2X wireless communication system, characterized in that, include: The status information receiving unit is used to receive the current status information of each user at each moment; The trajectory accuracy generation unit is used to extract the temporal features of the state information, and use the target deep neural network module to perform trajectory accuracy estimation processing for each user at the current time and the next time based on the temporal features, so as to obtain the trajectory accuracy at the current time and the trajectory accuracy at the next time. A communication resource scheduling unit is used to generate user communication priorities based on the current trajectory accuracy and the next trajectory accuracy, and to perform vehicle network communication resource scheduling according to the user communication priorities. The trajectory accuracy at the next moment includes the trajectory accuracy at the next moment after successful transmission and the trajectory accuracy at the next moment after transmission failure. The communication resource scheduling unit includes: The communication transmission success rate calculation unit is used to calculate the communication transmission success rate at the current moment. The revenue calculation unit is used to calculate the revenue of the dispatched vehicle at the current moment based on the communication transmission success rate, the trajectory accuracy at the current moment, the trajectory accuracy at the next moment after successful transmission, and the trajectory accuracy at the next moment after failed transmission, according to a first preset formula. The user communication priority calculation unit is used to calculate the user communication priority based on the revenue according to the second preset formula, and to perform vehicle network communication resource scheduling according to the user communication priority. The first preset formula and the second preset formula are as follows: ； ； in, The trajectory accuracy at the current moment is... The trajectory accuracy at the next moment after successful transmission. The trajectory accuracy at the next moment after the transmission failure. The communication transmission success rate, For the aforementioned benefits, for Users who are selected for priority scheduling at all times.

7. A computer device, characterized in that, It includes a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the vehicle network resource scheduling method for trajectory prediction as described in any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the vehicle network resource scheduling method for trajectory prediction as described in any one of claims 1 to 5.

Images