A method and related device for predicting and avoiding conflict at an intersection

Abstract

The present application discloses a method for predicting and avoiding conflicts at intersections and related devices, the method comprising obtaining characteristic information of the target vehicle when the target vehicle arrives at a set position corresponding to the intersection area; determining the predicted entry time and predicted exit time of the target vehicle based on the characteristic information; detecting conflicting vehicles based on the predicted entry time, the predicted exit time and the target time slot corresponding to the target vehicle; and changing the target time slot corresponding to the target vehicle when a conflicting vehicle is detected. The present application predicts whether vehicles using the same time slot will be in the intersection area at the same time based on the characteristic information of the target vehicle, and then determines whether there is a time slot conflict, and when there is a conflict, adjusts the target time slot occupied by the vehicle to reduce the time slot conflict at the intersection, thereby ensuring a low merge collision rate and optimizing network performance.

Description

Technical Field

[0001] The present application relates to the technical field of vehicle-mounted self-organizing network communication, and in particular to a method for predicting and avoiding conflicts at intersections and related devices. Background Art

[0002] Vehicular self-organizing networks have good situational awareness and can perceive potential dangers and abnormal traffic conditions outside the visual range. Therefore, they are widely regarded as an effective solution to improve traffic safety and efficiency. In the MAC protocol based on time division multiple access (TDMA), time is divided into frames, each of which has a fixed number of equal-length time slots. Vehicles synchronize time slots and access the channel once in each frame by occupying time slots. However, when two or more vehicles move together and occupy the same time slot, an access conflict occurs when two or more nodes two hops apart try to obtain the same available time slot; if two or more nodes that obtain the same time slot become members of the same two-hop set due to node activation or node movement, a merge conflict occurs.

[0003] At intersections, a large number of vehicles gather at intersections because of the convergence of traffic from multiple directions and the need for vehicles to wait for red lights, which can easily cause access conflicts and merge conflicts. However, existing conflict avoidance methods generally focus on time slot conflicts in two-way highway scenarios, but cannot effectively handle frequent time slot conflicts at intersections, resulting in the inability to obtain safety messages in a timely manner.

[0004] Therefore the prior art still needs to be improved and enhanced. Summary of the invention

[0005] The technical problem to be solved by the present application is to provide a method and related device for predicting and avoiding conflicts at intersections in view of the deficiencies of the prior art.

[0006] In order to solve the above technical problems, the first aspect of the embodiment of the present application provides a method for predicting and avoiding conflicts at an intersection, the method comprising:

[0007] When the target vehicle reaches a set position corresponding to the intersection area, characteristic information of the target vehicle is obtained, wherein the characteristic information includes vehicle information and signal light information;

[0008] Determining a predicted entry time and a predicted exit time of the target vehicle based on the characteristic information;

[0009] Detecting a conflicting vehicle based on the predicted entry time, the predicted exit time, and a target time slot corresponding to the target vehicle, wherein the conflicting vehicle occupies the target time slot;

[0010] When a conflicting vehicle is detected, the target time slot corresponding to the target vehicle is changed.

[0011] The method for predicting and avoiding conflicts at intersections, wherein the intersection area is an area constructed with an intersection as the center and a preset communication distance as the radius, wherein the set position is located outside the intersection area.

[0012] The method for predicting and avoiding conflicts at intersections, wherein the detecting whether there is a conflicting vehicle based on the predicted entry time, the predicted exit time, and the target time slot corresponding to the target vehicle specifically includes:

[0013] Acquire a candidate vehicle corresponding to the target vehicle according to the target time slot, wherein the candidate vehicle occupies the target time slot;

[0014] Detecting whether the candidate entry time of the candidate vehicle is within a time period formed by the predicted entry time and the predicted exit time;

[0015] If the candidate entry time is within the time period, the candidate vehicle is regarded as a conflicting vehicle.

[0016] The method for predicting and avoiding conflicts at intersections, wherein the traveling direction of the candidate vehicle is different from the traveling direction of the target vehicle.

[0017] The method for predicting and avoiding conflicts at intersections, wherein determining the predicted entry time and the predicted exit time of the target vehicle based on the characteristic information specifically includes:

[0018] Determine input information based on the feature information, and input the input information into a trained prediction network model;

[0019] Outputting the predicted entry time and predicted exit time of the target vehicle through the prediction network model;

[0020] Among them, the prediction network model includes a feature extraction module and a memory module, the feature extraction module is connected to the memory module, the memory module includes a memory unit and an attention unit, and the memory unit is connected to the attention unit.

[0021] The method for predicting and avoiding conflicts at intersections, wherein the step of determining input information based on the signal light information, the vehicle information, and the area information of the intersection area specifically includes:

[0022] Acquire characteristic information of each reference vehicle among a preset number of reference vehicles before the target vehicle;

[0023] The characteristic information of the target vehicle and the characteristic information of each referenced here are used as input information.

[0024] The method for predicting and avoiding conflicts at intersections, wherein the method further comprises:

[0025] The vehicles waiting for the traffic light corresponding to the intersection area are regarded as a vehicle queue, wherein the vehicles in the vehicle queue are located in the same lane and have the same driving direction;

[0026] The starting vehicle closest to the signal light and the ending vehicle farthest from the signal light in the vehicle queue are controlled to send beacons, and other vehicles in the vehicle queue stop sending beacons.

[0027] A second aspect of an embodiment of the present application provides a system for predicting and avoiding conflicts at an intersection, the system comprising:

[0028] An acquisition module, used for acquiring characteristic information of the target vehicle when the target vehicle reaches a set position corresponding to the intersection area, wherein the characteristic information includes vehicle information and signal light information;

[0029] A determination module, configured to determine a predicted entry time and a predicted exit time of the target vehicle based on the characteristic information;

[0030] A detection module, configured to detect a conflicting vehicle based on the predicted entry time, the predicted exit time, and a target time slot corresponding to the target vehicle, wherein the conflicting vehicle occupies the target time slot;

[0031] The changing module is used to change the target time slot corresponding to the target vehicle when a conflicting vehicle is detected.

[0032] A third aspect of an embodiment of the present application provides a computer-readable storage medium, which stores one or more programs. The one or more programs can be executed by one or more processors to implement the steps in any of the above-described methods for predicting and avoiding conflicts at intersections.

[0033] A fourth aspect of the embodiments of the present application provides a terminal device, comprising: a processor, a memory, and a communication bus; the memory stores a computer-readable program that can be executed by the processor;

[0034] The communication bus realizes the connection and communication between the processor and the memory;

[0035] When the processor executes the computer-readable program, the steps in any of the above-described methods for predicting and avoiding conflicts at intersections are implemented.

[0036] Beneficial effects: Compared with the prior art, the present application provides a method and related device for predicting and avoiding conflicts at intersections, the method comprising obtaining characteristic information of the target vehicle when the target vehicle reaches a set position corresponding to the intersection area; determining the predicted entry time and predicted exit time of the target vehicle based on the characteristic information; detecting conflicting vehicles based on the predicted entry time, the predicted exit time and the target time slot corresponding to the target vehicle; and changing the target time slot corresponding to the target vehicle when a conflicting vehicle is detected. The present application predicts whether vehicles using the same time slot will be in the intersection area at the same time based on the characteristic information of the target vehicle, and then determines whether there is a time slot conflict, and when there is a conflict, adjusts the target time slot occupied by the vehicle to reduce the time slot conflict at the intersection, thereby ensuring a low merge collision rate and optimizing network performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without inventive work.

[0038] Figure 1 This is a time slot division diagram for the method for predicting and avoiding conflicts at intersections provided in this application.

[0039] Figure 2 A schematic diagram of the application scenario.

[0040] Figure 3 This is a flow chart of the method for predicting and avoiding conflicts at intersections provided in this application.

[0041] Figure 4 This is a structural diagram of the prediction network model.

[0042] Figure 5 Schematic diagram of frequency adjustment for beacon transmission.

[0043] Figure 6 This is a schematic diagram of the structure of the intersection conflict prediction and avoidance system provided in this application.

[0044] Figure 7 This is a schematic diagram of the structure of the terminal device provided in this application. DETAILED DESCRIPTION

[0045] The present application provides a method and related device for predicting and avoiding conflicts at intersections. To make the purpose, technical solution and effect of the present application clearer and more specific, the present application is further described in detail with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0046] It will be understood by those skilled in the art that, unless expressly stated, the singular forms "one", "said", and "the" used herein may also include plural forms. It should be further understood that the term "comprising" used in the specification of the present application refers to the presence of the features, integers, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or groups thereof. It should be understood that when we refer to an element as being "connected" or "coupled" to another element, it may be directly connected or coupled to the other element, or there may be an intermediate element. In addition, the "connection" or "coupling" used herein may include wireless connection or wireless coupling. The term "and / or" used herein includes all or any unit and all combinations of one or more associated listed items.

[0047] It will be understood by those skilled in the art that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as those generally understood by those skilled in the art to which this application belongs. It should also be understood that terms such as those defined in general dictionaries should be understood to have meanings consistent with those in the context of the prior art, and will not be interpreted with idealized or overly formal meanings unless specifically defined as here.

[0048] It should be understood that the sequence numbers and sizes of the steps in this embodiment do not mean the order of execution. The execution order of each process is determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present application.

[0049] After research, it was found that the vehicle-mounted self-organizing network has good situational awareness and can perceive potential dangers and abnormal traffic conditions outside the visual range. Therefore, it is also widely regarded as an effective solution to improve traffic safety and traffic efficiency. In the MAC protocol based on time division multiple access (TDMA), time is divided into frames, each frame has a fixed number of equal-length time slots. The time slots are synchronized between vehicles and the channel is accessed once in each frame by occupying the time slots. However, when two or more vehicles move together and occupy the same time slot, an access conflict occurs when two or more nodes two hops apart try to obtain the same available time slot; if two or more nodes that obtain the same time slot become members of the same two-hop set due to node activation or node movement, a merge conflict occurs.

[0050] At intersections, a large number of vehicles gather at intersections because of the convergence of traffic from multiple directions and the need for vehicles to wait for red lights, which can easily cause access conflicts and merge conflicts. However, existing conflict avoidance methods generally focus on time slot conflicts in two-way highway scenarios, but cannot effectively handle frequent time slot conflicts at intersections, resulting in the inability to obtain safety messages in a timely manner.

[0051] In order to solve the above problems, in the embodiments of the present application.

[0052] The application content is further explained below through the description of embodiments in conjunction with the accompanying drawings.

[0053] This embodiment provides a method for predicting and avoiding conflicts at intersections. The time division method of the method is as follows: Figure 1 As shown, the time is divided into repeated frames of the same length, each frame includes a given number of time slots of fixed duration, wherein the time slots of each frame are divided into three non-overlapping time slot sets, respectively denoted as L set, R set and F set, L set is the time slots of the nodes of the left-hand-traveling vehicles, R set is the time slots of the nodes of the right-hand-traveling vehicles, and F set is the time slots of the RUS (Rode side unit) nodes, wherein RUS is a communication device installed on both sides of the road, responsible for broadcasting messages to surrounding vehicles. In this embodiment, the application scenario is as follows Figure 2 The intersection scenario, such as Figure 1 As shown, the left direction is one of the two intersecting directions of the intersection, and the right direction refers to one of the two intersecting directions of the intersection.

[0054] like Figure 3 As shown, the method includes:

[0055] S10. When the target vehicle reaches a set position corresponding to the intersection area, characteristic information of the target vehicle is obtained.

[0056] Specifically, the intersection area is an area centered on the intersection, for example, a circular area, a rectangular area, a square area, etc. centered on the intersection; the intersection area is the basis for restricting two vehicles from using the same time slot, that is, two vehicles in the intersection area do not use the same time slot to avoid merging conflicts in different directions. In one implementation, the intersection area can be a circular area constructed with the intersection as the center and the preset communication distance as the radius, wherein the preset communication distance is the communication distance of DSRC. At the same time, let Pi n is the position where the vehicle enters the intersection, P out is the position where the vehicle leaves the intersection, P pd is the setting position where the RSU starts to collect the information required for prediction, P pd Outside the intersection area, when the vehicle is in Ppd When the vehicle is in the state of motion, the RSU starts to collect vehicle information and traffic light information.

[0057] Further, if Figure 2 As shown, the intersection area is the dotted box area in the figure, P center Represents the center of the intersection, which is also the location of the traffic light. If vehicle A enters from the If you enter the intersection area, you can leave the intersection from four directions. The four exit locations are recorded as represents leaving from the four directions of east, south, west and north. After P pd When P pd Start collecting vehicle information and traffic light information so that the vehicle A can predict the vehicle A’s input from the entry position based on the collected feature information. Enter the time to enter the intersection area and the time when the vehicle exits the intersection area

[0058] In one implementation, the feature information includes vehicle information and signal light information, wherein the signal light information is obtained by collecting information about signal lights at intersections in the intersection area. The vehicle information may include the vehicle type, the speed at which the vehicle passes through a set position, the acceleration at which the vehicle passes through a set position, the angle at which the vehicle travels, and the direction of the vehicle at the intersection; the signal light information includes the signal light phase when the vehicle passes through a set position, the time to switch to the next phase, and the traffic flow density.

[0059] Based on this, the characteristic information can be expressed as:

[0060] x i =[type i ,speed i ,acc i ,angle i , direction i , phase i , TRSNP i , VFD i ]

[0061] Among them, type i Represents the vehicle type of vehicle i, speed i With acc i Represents vehicle i passing through P pd The velocity and acceleration at angle i Represents the angle and direction of vehicle i when it is traveling i represents the turning of vehicle i at the intersection, phasei TRSNP i Represents vehicle i passing through P pd The signal light phase and the time to switch to the next phase, VFD i Represents the traffic flow density on the road when vehicle i is traveling.

[0062] S20. Determine a predicted entry time and a predicted exit time of the target vehicle based on the characteristic information.

[0063] Specifically, the predicted entry time is the time when the target vehicle enters the intersection area, and the predicted exit time is the time when the target vehicle exits the intersection area. The predicted entry time and the predicted exit time are predicted by a preset network model based on feature information; they can also be calculated by a traditional method based on feature information, wherein the feature information can be input into the prediction network model through the process of prediction by the preset network model, and the predicted entry time and predicted exit time can be output by the prediction network model; the calculation by the traditional method can predict the first driving time of the vehicle in the intersection area and the second driving time of the vehicle from the set position to the entry position through the vehicle speed, acceleration, traffic light information, etc., and the predicted entry time and predicted exit time are determined based on the first driving time, the second driving time and the time when the vehicle passes the set position.

[0064] In one implementation, determining the predicted entry time and the predicted exit time of the target vehicle based on the feature information specifically includes:

[0065] Determine input information based on the feature information, and input the input information into a trained prediction network model;

[0066] The predicted entry time and the predicted exit time of the target vehicle are outputted through the prediction network model.

[0067] Specifically, the prediction network model is pre-trained, and after the feature information is input into the prediction network model, the prediction network model can output the predicted entry time and predicted exit time of the target vehicle. Figure 4 As shown, the prediction network model includes a feature extraction module and a memory module, the feature extraction module is connected to the memory module, the memory module includes a memory unit and an attention unit, and the memory unit is connected to the attention unit.

[0068] The feature extraction module is used to extract vehicle kinematic features and signal light phase features, wherein the feature extraction module may use a convolutional neural network, such as a CNN network. In this embodiment, the feature extraction module may include a number of convolution units, each of which may include a convolution layer, a pooling layer, and a fully connected layer. Since the phase of the signal light, the time to switch to the next phase, and the traffic density are all correlated with the time when the vehicle passes through the intersection, a convolutional neural network is used as a feature extraction module to extract high-dimensional spatial features of feature information under the action of the convolution kernel, wherein the high-dimensional spatial features can be expressed as:

[0069] y t =σ cnn (W t *x t +b t )

[0070] Among them, y t represents the output value after convolution, σ cnn represents the activation function, W t is the weight matrix, * is the convolution operation, x t is the input vector, b t For bias.

[0071] The memory unit is used to capture and learn the time dependency of the time when the vehicle passes through the intersection. The attention unit is configured with an Attention mechanism, which adjusts the weight of the current input to highlight the role of the input that has a significant impact on the prediction result. In this embodiment, the memory unit can use BiLSTM, which has two independent LSTMs, a forward LSTM and a backward LSTM; during the calculation process, the two networks will calculate the forward hidden sequence respectively. and the backward hidden sequence The Attention mechanism is for each h output of the n activation layers of BiLSTM i Assign a weight α i , and meet The output of the attention unit is a vector c, which is obtained by multiplying the attention weight by the hidden output, expressed as: c = [α1h1, α2h2, …, α n h n ]; then predict the predicted entry time and the predicted exit time based on the output items of the attention unit.

[0072] Furthermore, LSTM contains three gates, namely input gate i t 、Forget gate t and output gate o t , where the input gate determines the tThe proportion of vehicle and signal information in the LSTM that is allowed to be updated at time t, the forget gate determines the degree of influence of the previous vehicle and signal status on the current vehicle status; the output gate determines the proportion of information that should pass through the memory state at each time step. The LSTM calculation process is as follows:

[0073] i t =σ(W i x t +U i h t-1 )

[0074] f t =σ(W f x t +U f h t-1 )

[0075] o t =σ(W o x t +U o h t-1 )

[0076]

[0077]

[0078]

[0079] Among them, i t 、f t and t They represent the input gate, forget gate and input gate respectively, h t represents the output of LSTM, represents the Hadamard product, the matrix W i , U i , W f , U f , W o and U o Associated with the input gate, output gate and forget gate respectively.

[0080] In one implementation, in order to increase the amount of information obtained by the prediction network model, the feature information of a preset number of vehicles before the target vehicle can also be used as input information of the preset network. Based on this, the input information is determined based on the signal light information, the vehicle information and the area information of the intersection area, specifically including:

[0081] Acquire characteristic information of each reference vehicle among a preset number of reference vehicles before the target vehicle;

[0082] The characteristic information of the target vehicle and the characteristic information of each referenced here are used as input information.

[0083] Specifically, the preset number can be pre-set, and the reference vehicle is located before the target vehicle in the time sequence and is continuous in the time sequence. It can be understood that the preset number of reference vehicles is a preset number of vehicles that are located before the target vehicle in the time sequence and are continuous, and the preset number of reference vehicles are located at the last vehicle in the time sequence and are adjacent to the target vehicle. In addition, the information content included in the characteristic information of the reference vehicle is the same as the characteristic information included in the characteristics of the target vehicle. Therefore, the input information can be expressed as:

[0084] X i =[x i _ N , x i _ N+1 , …, x i ]

[0085] Among them, x i represents the characteristic information of vehicle i, x i-N represents the characteristic information of the reference vehicle iN, x i-N+1 Represents the characteristic information of reference vehicle i-N+1, where N represents a preset number.

[0086] Furthermore, in practical applications, the prediction network model needs to be pre-trained. The model can predict the results more accurately by training on the training set, wherein the training set is the historical data collected at the intersection, including the input data X i and Y as the predicted value i in and Y i out , the size of the training set can be around 20,000.

[0087] S30. Detecting a conflicting vehicle based on the predicted entry time, the predicted exit time, and a target time slot corresponding to the target vehicle, wherein the conflicting vehicle occupies the target time slot.

[0088] Specifically, the detected conflicting vehicle is a vehicle occupying the target time slot, and the driving direction at the intersection of the intersection area intersects with the driving direction of the target vehicle at the intersection. In other words, the conflicting vehicle is a vehicle that has a merge conflict with the target vehicle in the intersection area. In this embodiment, after obtaining the predicted entry time and the predicted exit time, the time period of the target vehicle passing through the intersection area can be determined, and it can be detected whether there is a vehicle occupying the target time slot entering the intersection area during the time period to detect whether there is an obstacle vehicle, that is, a vehicle occupying the target time slot and entering the intersection area during the time period corresponding to the target vehicle is regarded as an obstacle vehicle, wherein the time period is formed by the predicted entry time and the predicted exit time.

[0089] Based on this, the detecting whether there is a conflicting vehicle based on the predicted entry time, the predicted exit time and the target time slot corresponding to the target vehicle specifically includes:

[0090] S31, acquiring a candidate vehicle corresponding to the target vehicle according to the target time slot, wherein the candidate vehicle occupies the target time slot;

[0091] S32, detecting whether the candidate entry time of the candidate vehicle is within a time period formed by the predicted entry time and the predicted exit time;

[0092] S33: If the candidate entry time is within the time period, the candidate vehicle is regarded as a conflicting vehicle.

[0093] Specifically, when a vehicle passes through a set position, a record R is created for the vehicle. The record R includes the vehicle passing through P pd Time t p , predicted entry time t in , predicted departure time t out , vehicle ID and target time slot s used by the vehicle i , when the vehicle passes P pd When the vehicle leaves the intersection, the record R of the vehicle is stored. When the vehicle changes the time slot used, the target time slot s in the record is adaptively modified. i Therefore, when vehicle i passes P pd At point s, the target time slot s can be determined by the records R of each vehicle. i Is it used by other vehicles? If vehicle j also uses the target time slot s i When the vehicle j is used as a candidate vehicle, the predicted entry time of the vehicle j is used as the candidate entry time, and then it is detected whether the candidate entry time is within the time period formed by the predicted entry time and the predicted exit time, that is, the detection Is it satisfied? in, It indicates the predicted entry time. Indicates the predicted departure time. represents the candidate entry time; satisfy , vehicle j is regarded as a conflicting vehicle.

[0094] S40: When a conflicting vehicle is detected, the target time slot corresponding to the target vehicle is changed.

[0095] Specifically, when a conflicting vehicle is detected, it means that when the target vehicle uses the target time slot, a merge collision may occur in the intersection area. Therefore, it is necessary to adjust the target time slot occupied by the target vehicle to avoid the merge collision. Of course, it is worth noting that after changing the target time slot corresponding to the target vehicle, it is still judged whether the target vehicle has a merge collision problem. The judgment process is the same as the above process, which will not be repeated here.

[0096] In one implementation, the method further includes:

[0097] The vehicles waiting for the traffic light corresponding to the intersection area are regarded as a vehicle queue, wherein the vehicles in the vehicle queue are located in the same lane and have the same driving direction;

[0098] The starting vehicle closest to the signal light and the ending vehicle farthest from the signal light in the vehicle queue are controlled to send beacons, and other vehicles in the vehicle queue stop sending beacons.

[0099] Specifically, Figure 5 As shown in the figure, when vehicles are waiting for the traffic light, they wait for the red light in their respective lanes in turn. The vehicles in each lane are regarded as a vehicle queue, and the vehicle closest to the traffic light is the head of the queue, which is recorded as The one farthest from the signal light is the tail of the queue, denoted by When vehicles are waiting for a red light, only the head and tail vehicles in the queue are allowed to send beacons, and the vehicles in the middle of the queue stop sending beacons to save time slot resources and reduce time slot conflicts.

[0100] In one implementation, only the head and tail vehicles in the queue are allowed to send beacons, and the vehicles in the middle of the queue stop sending beacons, which can be determined according to the beacon sending coefficient S. That is, each vehicle in the vehicle queue will obtain a beacon sending coefficient S, and determine whether to send a beacon according to the beacon sending coefficient S. When S=0, it means that the vehicle is in the middle of the queue, the vehicle speed is 0, and t m The signal light will not switch within seconds. At this time, all vehicles in the vehicle queue except the front vehicle and the last vehicle will stop sending beacons. Conversely, when S≠0, each vehicle in the vehicle queue will still send beacons at a normal frequency.

[0101] In this embodiment, the expression of the sending beacon coefficient S can be:

[0102]

[0103] where j = 0, 1, ..., j r is the vehicle number in the queue, the first vehicle is numbered 0, and the last vehicle is numbered j r v represents the vehicle speed, ReLU is the activation function, and t mIndicates the remaining time for the signal to switch to the next stage. Before the signal switching stage, the vehicle will resume sending beacons, and the advance time is marked as t s .

[0104] In summary, this embodiment provides a method and related device for predicting and avoiding conflicts at intersections, wherein the method includes obtaining characteristic information of the target vehicle when the target vehicle arrives at a set position corresponding to the intersection area; determining the predicted entry time and predicted exit time of the target vehicle based on the characteristic information; detecting conflicting vehicles based on the predicted entry time, the predicted exit time and the target time slot corresponding to the target vehicle; and changing the target time slot corresponding to the target vehicle when a conflicting vehicle is detected. This embodiment predicts the predicted entry time and predicted exit time of the target vehicle through the prediction network model CNN-BiLSTM-Attention, wherein the prediction network model takes into account the time when the vehicle waits for the red light at the intersection and the vehicle will leave the intersection from different directions. The model can accurately predict the time required for the vehicle to pass through the intersection by learning historical data, which can effectively avoid intersection merge conflicts. In addition, this embodiment can further save time slot resources and reduce time slot conflicts by adjusting the beacon transmission frequency when the vehicle waits for the red light.

[0105] Based on the above-mentioned intersection conflict prediction and avoidance method, this embodiment provides an intersection conflict prediction and avoidance system, such as Figure 6 As shown, the system comprises:

[0106] The acquisition module 100 is used to acquire characteristic information of the target vehicle when the target vehicle arrives at a set position corresponding to the intersection area, wherein the characteristic information includes vehicle information and signal light information;

[0107] A determination module 200, configured to determine a predicted entry time and a predicted exit time of the target vehicle based on the characteristic information;

[0108] A detection module 300, configured to detect a conflicting vehicle based on the predicted entry time, the predicted exit time, and a target time slot corresponding to the target vehicle, wherein the conflicting vehicle occupies the target time slot;

[0109] The changing module 400 is used to change the target time slot corresponding to the target vehicle when a conflicting vehicle is detected.

[0110] Based on the above-mentioned method for predicting and avoiding conflicts at intersections, this embodiment provides a computer-readable storage medium, which stores one or more programs. The one or more programs can be executed by one or more processors to implement the steps in the method for predicting and avoiding conflicts at intersections as described in the above-mentioned embodiment.

[0111] Based on the above-mentioned intersection conflict prediction and avoidance method, the present application also provides a terminal device, such as Figure 7 As shown, it includes at least one processor (processor) 20; display screen 21; and memory (memory) 22, and may also include a communications interface (Communications Interface) 23 and a bus 24. Among them, the processor 20, the display screen 21, the memory 22 and the communication interface 23 can communicate with each other through the bus 24. The display screen 21 is configured to display a preset user guide interface in the initial setting mode. The communication interface 23 can transmit information. The processor 20 can call the logic instructions in the memory 22 to execute the method in the above embodiment.

[0112] In addition, the logic instructions in the memory 22 can be implemented in the form of software functional units and can be stored in a computer-readable storage medium when sold or used as an independent product.

[0113] The memory 22 is a computer-readable storage medium that can be configured to store software programs, computer executable programs, such as program instructions or modules corresponding to the methods in the embodiments of the present disclosure. The processor 20 executes functional applications and data processing by running the software programs, instructions or modules stored in the memory 22, that is, implementing the methods in the above embodiments.

[0114] The memory 22 may include a program storage area and a data storage area, wherein the program storage area may store an operating system and at least one application required for a function; the data storage area may store data created according to the use of the terminal device, etc. In addition, the memory 22 may include a high-speed random access memory and may also include a non-volatile memory. For example, a variety of media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a disk or an optical disk, may also be a transient storage medium.

[0115] In addition, the specific process of loading and executing the multiple instruction processors in the above-mentioned storage medium and the terminal device has been described in detail in the above-mentioned method, and will not be described one by one here.

[0116] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for predicting and avoiding conflicts at intersections, characterized in that: The method comprises: When the target vehicle reaches a set position corresponding to the intersection area, characteristic information of the target vehicle is obtained, wherein the characteristic information includes vehicle information and signal light information; Determining a predicted entry time and a predicted exit time of the target vehicle based on the characteristic information; Detecting a conflicting vehicle based on the predicted entry time, the predicted exit time, and a target time slot corresponding to the target vehicle, wherein the conflicting vehicle occupies the target time slot; When a conflicting vehicle is detected, the target time slot corresponding to the target vehicle is changed.

2. The method for predicting and avoiding conflicts at intersections according to claim 1, characterized in that: The intersection area is an area constructed with the intersection as the center and the preset communication distance as the radius, wherein the set position is located outside the intersection area.

3. The method for predicting and avoiding conflicts at intersections according to claim 1, characterized in that: The detecting whether there is a conflicting vehicle based on the predicted entry time, the predicted exit time, and the target time slot corresponding to the target vehicle specifically includes: Acquire a candidate vehicle corresponding to the target vehicle according to the target time slot, wherein the candidate vehicle occupies the target time slot; Detecting whether the candidate entry time of the candidate vehicle is within a time period formed by the predicted entry time and the predicted exit time; If the candidate entry time is within the time period, the candidate vehicle is regarded as a conflicting vehicle.

4. The method for predicting and avoiding conflicts at intersections according to claim 3, characterized in that: The traveling direction of the candidate vehicle is different from the traveling direction of the target vehicle.

5. The method for predicting and avoiding conflicts at intersections according to claim 1, characterized in that: The step of determining the predicted entry time and the predicted exit time of the target vehicle based on the characteristic information specifically includes: Determine input information based on the feature information, and input the input information into a trained prediction network model; Outputting the predicted entry time and predicted exit time of the target vehicle through the prediction network model; Among them, the prediction network model includes a feature extraction module and a memory module, the feature extraction module is connected to the memory module, the memory module includes a memory unit and an attention unit, and the memory unit is connected to the attention unit.

6. The method for predicting and avoiding conflicts at intersections according to claim 5, characterized in that: The determining of the input information based on the signal light information, the vehicle information and the area information of the intersection area specifically includes: Acquire characteristic information of each reference vehicle among a preset number of reference vehicles before the target vehicle; The characteristic information of the target vehicle and the characteristic information of each referenced here are used as input information.

7. The method for predicting and avoiding conflicts at intersections according to claim 1, characterized in that: The method further comprises: The vehicles waiting for the traffic light corresponding to the intersection area are regarded as a vehicle queue, wherein the vehicles in the vehicle queue are located in the same lane and have the same driving direction; The starting vehicle closest to the signal light and the ending vehicle farthest from the signal light in the vehicle queue are controlled to send beacons, and other vehicles in the vehicle queue stop sending beacons.

8. A conflict prediction and avoidance system at an intersection, characterized in that: The system comprises: An acquisition module, used for acquiring characteristic information of the target vehicle when the target vehicle reaches a set position corresponding to the intersection area, wherein the characteristic information includes vehicle information and signal light information; A determination module, configured to determine a predicted entry time and a predicted exit time of the target vehicle based on the characteristic information; A detection module, configured to detect a conflicting vehicle based on the predicted entry time, the predicted exit time, and a target time slot corresponding to the target vehicle, wherein the conflicting vehicle occupies the target time slot; The changing module is used to change the target time slot corresponding to the target vehicle when a conflicting vehicle is detected.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the steps in the method for predicting and avoiding conflicts at intersections as described in any one of claims 1-7.

10. A terminal device, characterized in that: include: Processor, memory and communication bus; The memory stores a computer-readable program executable by the processor; The communication bus realizes the connection and communication between the processor and the memory; When the processor executes the computer-readable program, the steps of the method for predicting and avoiding conflicts at intersections as described in any one of claims 1 to 7 are implemented.

Images