Vehicle track prediction method and device, vehicle and storage medium

Abstract

The invention discloses a vehicle track prediction method and device, a vehicle and a storage medium, and the method comprises the steps: obtaining the position information of a test vehicle; determining a target dangerous vehicle according to the position information; obtaining vehicle driving information and a collision point type of the target dangerous vehicle; and inputting the vehicle driving information and the collision point type into the target trajectory prediction model to obtain the driving trajectory of the target dangerous vehicle. The technical problem that the driving track of the dangerous target cannot be accurately predicted due to the fact that the motion direction of the dangerous target is not considered when the dangerous target is detected during vehicle driving in the prior art is solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of automation control, in particular to a vehicle trajectory prediction method and device, a vehicle and a storage medium. BACKGROUND

[0002] In the field of intelligent driving assistance systems, especially for dangerous identification and early warning technology in the process of vehicle driving, the existing scheme mainly adopts two methods to identify potential dangerous targets: one is to rely on lane line information to define the dangerous source in the vehicle lane; the second is to evaluate the potential collision target based on the real-time motion state of the vehicle combined with the predicted driving trajectory. However, these methods often rely too much on static lane lines or single vehicle trajectory prediction, and do not fully consider the dynamic changes of targets in the traffic environment and the influence of their motion direction on the danger level.

[0003] Specifically in application scenarios, intelligent driving systems are difficult to effectively deal with when dealing with complex road conditions, such as multi-lane environments, intersections or lane changing situations. Due to the inability to distinguish the difference in danger posed by traffic targets in different directions, the system is prone to false positives or false negatives when predicting target behavior, especially when the target quickly changes lanes, approaches or crosses the driving path of the vehicle. This leads to instability and decreased accuracy of the warning system, especially in high traffic density and non-linear motion scenarios, the system may frequently appear false alarms or missed alarms, seriously interfering with the driver's decision-making, reducing the driving experience and safety.

[0004] In addition, the existing technology lacks consideration of road curvature and potential lane changing intentions when dealing with curves and special intersections, further exacerbating the limitations of the warning system, making the effectiveness and reliability of the system in complex road conditions greatly discounted.

[0005] Therefore, there is an urgent need for an innovative vehicle trajectory prediction method to address the shortcomings of existing technology. SUMMARY

[0006] The embodiments of the present application provide a vehicle trajectory prediction method, device, vehicle and storage medium to at least solve the technical problem that the existing technology cannot accurately predict the driving trajectory of the dangerous target when detecting the dangerous target during vehicle driving, because the motion direction of the dangerous target is not considered.

[0007] According to an embodiment of the present application, a vehicle trajectory prediction method is provided, comprising: obtaining position information of a test vehicle; determining a target dangerous vehicle according to the position information; obtaining vehicle driving information and a collision point type of the target dangerous vehicle; inputting the vehicle driving information and the collision point type into a target trajectory prediction model to obtain a driving trajectory of the target dangerous vehicle.

[0008] Optionally, the method further comprises: determining a first dangerous area and a second dangerous area according to the position information, wherein the second dangerous area is larger than the first dangerous area; determining a plurality of initial dangerous vehicles according to the first dangerous area and the second dangerous area; calculating a plurality of collision durations of the plurality of initial dangerous vehicles, wherein each initial dangerous vehicle corresponds to a collision duration; and determining a target dangerous vehicle from the plurality of initial dangerous vehicles according to the plurality of collision durations.

[0009] Optionally, the method further comprises: obtaining a plurality of first initial dangerous vehicles in the first dangerous area; obtaining a plurality of second initial dangerous vehicles in the second dangerous area; calculating a plurality of dangerous collision durations of the plurality of second initial dangerous vehicles, wherein each second initial dangerous vehicle corresponds to a dangerous collision duration; determining a plurality of second dangerous vehicles from the plurality of second initial dangerous vehicles according to the plurality of dangerous collision durations; and determining the plurality of initial dangerous vehicles according to the plurality of first initial dangerous vehicles and the plurality of second dangerous vehicles.

[0010] Optionally, the method further comprises: comparing the plurality of dangerous collision durations with a preset collision duration to obtain a first comparison result; and determining the plurality of second dangerous vehicles according to the first comparison result, wherein a second dangerous vehicle is a vehicle with a dangerous collision duration greater than the preset collision duration.

[0011] Optionally, the method further comprises: constructing a two-dimensional coordinate system according to the position information, wherein an origin of the two-dimensional coordinate system is a center point of the test vehicle; determining an initial two-dimensional coordinate of each initial dangerous vehicle in the two-dimensional coordinate system; obtaining a speed component of each initial dangerous vehicle during driving, wherein the speed component includes a speed longitudinal component and a speed transverse component; and calculating a collision duration of each initial dangerous vehicle according to the initial two-dimensional coordinate, the speed component, and a vehicle body width value of each initial dangerous vehicle.

[0012] Optionally, the method further comprises: calculating a coordinate assignment of a collision point according to the initial two-dimensional coordinate and the speed component; comparing the coordinate assignment with a first preset value and a second preset value respectively to obtain a second comparison result, wherein the second preset value is greater than the first preset value; and determining a collision point type according to the second comparison result, wherein the collision point type includes a front cross collision, a side cross collision, and a rear cross collision.

[0013] Optionally, the method further comprises: determining a driving environment of the test vehicle; and determining a target trajectory prediction model from a plurality of trajectory prediction models according to the driving environment.

[0014] According to one of the embodiments of the present application, a vehicle trajectory prediction device is provided, comprising: a first obtaining module configured to obtain position information of a test vehicle; a first determining module configured to determine a target dangerous vehicle according to the position information; a second obtaining module configured to obtain vehicle driving information and a collision point type of the target dangerous vehicle; and an input module configured to input the vehicle driving information and the collision point type into a target trajectory prediction model to obtain a driving trajectory of the target dangerous vehicle.

[0015] Optionally, the first determining module comprises: a first determining unit configured to determine a first dangerous area and a second dangerous area according to the position information, wherein the second dangerous area is larger than the first dangerous area; a second determining unit configured to determine a plurality of initial dangerous vehicles according to the first dangerous area and the second dangerous area; and a first calculating unit configured to calculate a plurality of collision durations of the plurality of initial dangerous vehicles, wherein each initial dangerous vehicle corresponds to one collision duration; and a third determining unit configured to determine the target dangerous vehicle from the plurality of initial dangerous vehicles according to the plurality of collision durations.

[0016] Optionally, the second determining unit comprises: a first obtaining sub-unit configured to obtain a plurality of first dangerous vehicles in the first dangerous area; a second obtaining sub-unit configured to obtain a plurality of second initial dangerous vehicles in the second dangerous area; a first calculating sub-unit configured to calculate a plurality of dangerous collision durations of the plurality of second initial dangerous vehicles, wherein each second initial dangerous vehicle corresponds to one dangerous collision duration; a first determining sub-unit configured to determine a plurality of second dangerous vehicles from the plurality of second initial dangerous vehicles according to the plurality of dangerous collision durations; and a second determining sub-unit configured to determine the plurality of initial dangerous vehicles according to the plurality of first dangerous vehicles and the plurality of second dangerous vehicles.

[0017] Optionally, the first determining sub-unit comprises: a comparing sub-unit configured to compare the plurality of dangerous collision durations with a preset collision duration to obtain a first comparison result; and a third determining sub-unit configured to determine the plurality of second dangerous vehicles according to the first comparison result, wherein the second dangerous vehicle is a vehicle with a dangerous collision duration greater than the preset collision duration.

[0018] Optionally, the first calculating unit comprises: a constructing sub-unit configured to construct a two-dimensional coordinate system according to the position information, wherein the origin of the two-dimensional coordinate system is the center point of the test vehicle; a fourth determining sub-unit configured to determine an initial two-dimensional coordinate of each initial dangerous vehicle in the two-dimensional coordinate system; a third obtaining sub-unit configured to obtain a speed component of each initial dangerous vehicle during driving, wherein the speed component comprises a speed longitudinal component and a speed transverse component; and a second calculating sub-unit configured to calculate the collision duration of each initial dangerous vehicle according to the initial two-dimensional coordinate, the speed component, and a vehicle body width value of each initial dangerous vehicle.

[0019] Optionally, the second obtaining module comprises: a second calculation unit, configured to calculate a coordinate assignment of the collision point according to the initial two-dimensional coordinate and the velocity component; a comparison unit, configured to compare the coordinate assignment with a first preset value and a second preset value respectively to obtain a second comparison result, wherein the second preset value is greater than the first preset value; and a fourth determination unit, configured to determine the collision point type according to the second comparison result, wherein the collision point type comprises a forward crossing collision, a lateral crossing collision and a rearward crossing collision.

[0020] Optionally, the vehicle trajectory prediction apparatus further comprises: a second determination module, configured to determine a driving environment of the test vehicle; and a third determination module, configured to determine the target trajectory prediction model from the plurality of trajectory prediction models according to the driving environment.

[0021] According to an embodiment of the present application, a vehicle is further provided, comprising a memory and a processor, the memory stores a computer program, and the processor is configured to execute the computer program to perform the vehicle trajectory prediction method in any one of the above.

[0022] According to an embodiment of the present application, an electronic device is further provided, comprising a memory and a processor, the memory stores a computer program, and the processor is configured to execute the computer program to perform the vehicle trajectory prediction method in any one of the above.

[0023] According to an embodiment of the present application, a non-volatile storage medium is further provided, the non-volatile storage medium stores a computer program, and the computer program is configured to perform the vehicle trajectory prediction method in any one of the above when executed.

[0024] According to an embodiment of the present application, a computer program product is further provided, the computer program product stores a computer program, and the computer program is configured to perform the steps of the vehicle trajectory prediction method in any one of the above when executed by a processor.

[0025] In the embodiments of the present application, the position information of the test vehicle is obtained, and the target dangerous vehicle is determined according to the position information, so as to achieve the purpose of obtaining the vehicle driving information and the collision point type of the target dangerous vehicle, thereby achieving the technical effect of inputting the vehicle driving information and the collision point type into the target trajectory prediction model to obtain the driving trajectory of the target dangerous vehicle, and further solving the technical problem that in the prior art, when a dangerous target is detected during vehicle driving, the movement direction of the dangerous target is not considered, so that the driving trajectory of the dangerous target cannot be accurately predicted. BRIEF DESCRIPTION OF DRAWINGS

[0026] The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiments of the application and together with the description serve to explain the application. In the drawings:

[0027] Figure 1 is a flow chart of a method for predicting a trajectory of a vehicle according to an embodiment of the present application;

[0028] Figure 2 is a schematic diagram of a test vehicle and a dangerous vehicle according to an embodiment of the present application;

[0029] Figure 3 is a schematic diagram of a dangerous vehicle trajectory prediction according to an embodiment of the present application;

[0030] Figure 4 is a structural block diagram of a device for predicting a trajectory of a vehicle according to an embodiment of the present application. DETAILED DESCRIPTION

[0031] In order to make the technical personnel of the present application better understand the present application, the following will be combined with the drawings of the embodiments of the present application, and the technical solutions in the embodiments of the present application will be described clearly and completely. Obviously, the described embodiments are only a part of the embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor should belong to the scope of protection of the present application.

[0032] It should be noted that the terms "first", "second", and the like in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and do not necessarily indicate a specific order or a chronological sequence. It should be understood that the terms thus used can be interchanged under appropriate circumstances, so that the embodiments of the present application described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device including a series of steps or units does not necessarily limit to those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0033] According to an embodiment of the present application, an embodiment of a method for predicting a trajectory of a vehicle is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system comprising at least one set of computer executable instructions, and although a logical sequence is shown in the flowchart, in some cases, the steps shown or described herein can be executed in an order different from that described herein.

[0034] The method embodiments can also be implemented in an electronic device comprising a memory and a processor, a similar control device, or a vehicle terminal. Taking the vehicle terminal as an example, the vehicle terminal can include one or more processors and a memory for storing data. Optionally, the vehicle terminal described above can also include a communication device for communication function and a display device. Those skilled in the art can understand that the above structural description is only illustrative, and does not limit the structure of the vehicle terminal described above. For example, the vehicle terminal can also include more or less components than the above structural description, or have a different configuration from the above structural description.

[0035] The processor can include one or more processing units. For example, the processor can include processing devices such as a central processing unit (CPU), a graphics processing unit (GPU), a digital signal processing (DSP) chip, a microprocessor, a programmable logic device (FPGA), a neural-network processing unit (NPU), a tensor processing unit (TPU), an artificial intelligent (AI) type processor, and the like. Different processing units can be independent components or integrated into one or more processors. In some examples, the electronic device can also include one or more processors.

[0036] The memory can be used to store a computer program, for example, a computer program corresponding to the vehicle trajectory prediction method in the embodiments of the present application. The processor implements the vehicle trajectory prediction method described above by running the computer program stored in the memory. The memory can include a high-speed random access memory, and can also include a non-volatile memory, such as one or more magnetic storage devices, flash memories, or other non-volatile solid-state memories. In some examples, the memory can further include a memory remotely disposed relative to the processor, which can be connected to the electronic device through a network. Examples of the network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

[0037] The communication device is configured to receive or transmit data via a mesh network. The mesh network can include, for example, a wireless mesh network provided by a communication service provider of the mobile terminal. In one example, the communication device includes a network interface controller (NIC) that is configured to connect to other mesh devices via a base station to enable communication with the Internet. In one example, the communication device can include a radio frequency (RF) module that is configured to communicate with the Internet via a wireless connection. In some embodiments, the communication device is configured to connect to a mobile device, such as a smartphone or tablet, and can be configured to transmit instructions to the vehicle terminal via the mobile device.

[0038] The display device can be a touch screen liquid crystal display (LCD) and a touch display (also referred to as a "touch screen" or "touch display screen"). The liquid crystal display can enable a user to interact with a user interface of the vehicle terminal. In some embodiments, the vehicle terminal has a graphical user interface (GUI) that a user can interact with via finger contacts and / or gestures on a touch-sensitive surface. The human-machine interaction function can include a gear shifting function of the vehicle, and executable instructions for performing the human-machine interaction function are configured / stored in one or more computer program products or readable storage media executable by a processor.

[0039] Figure 1 is a flowchart of a method for predicting a trajectory of a vehicle according to an embodiment of the present application, as shown in Figure 1 The method comprises the following steps:

[0040] In step S102, position information of the test vehicle is obtained.

[0041] Optionally, the execution subject of the present embodiment is a trajectory prediction system. It should be noted that other electronic devices and processors can also be used as the execution subject, which is not limited herein.

[0042] In the technical solution provided in step S102 of the present application, the latitude and longitude coordinates of the ground receiver can be determined by a satellite positioning system, i.e., by receiving signals from multiple satellites. The satellite positioning system is based on the principle of triangulation, and the precise geographic coordinates of the vehicle are determined by calculating the distances between the receiver and at least four satellites, using time delay or frequency difference as a distance measurement method.

[0043] Specifically, the satellite positioning system determines the geographic coordinates of a specific location on the earth's surface by receiving satellite signals, and is widely used in navigation, positioning and speed measurement fields.

[0044] As another alternative implementation, the inertial measurement unit contains an accelerometer and a gyroscope for measuring the acceleration, angular velocity and attitude change of the vehicle. By continuously integrating these measurements, the relative displacement of the vehicle can be calculated. The odometer is a sensor that calculates the distance traveled based on the rotation information of the vehicle tires. By combining the inertial measurement unit data with the odometer information, high-precision position updates can be provided in a short time, especially in cases where satellite signals are poor (such as tunnels or high-rise areas), as a supplement to GPS positioning, maintaining the continuity of positioning.

[0045] It is worth noting that whether through satellite positioning system or in combination with inertial measurement unit and odometer, obtaining the position information of the test vehicle can ensure that the intelligent driving assistance system has the ability to track the position of the vehicle in real time. Thus it can ensure that the early warning system can accurately grasp the position of the vehicle at any point in time, so as to make decisions based on actual driving conditions when target screening and prediction. Further, accurate position information also helps to improve the modeling accuracy of the road environment, enhance the vehicle's perception of external environmental changes, and realize functions such as planning obstacle avoidance routes and automated driving.

[0046] Step S104, determining the target dangerous vehicle according to the position information.

[0047] In the technical solution provided by the above step S104 of the present application, first, as shown in Figure 2 The system divides the first dangerous area by the preset requirement, and the vehicles in the first dangerous area are all initial dangerous vehicles. Further, for the vehicles outside the first area, a second area needs to be divided, which is slightly larger than the first area. The collision time of the vehicles in the second area is calculated, and the obtained multiple collision times are compared with the set threshold value, and then the vehicles with collision time less than the set threshold value are also regarded as initial dangerous vehicles.

[0048] Further, the collision time of the above multiple initial dangerous vehicles is calculated again, and the vehicle with the least collision time is selected as the target dangerous vehicle.

[0049] Specifically, the time to collision (TTC) is an algorithm that predicts the time required for a target object and a test vehicle to collide by calculating the relative motion between them, that is, the collision time is an important indicator for evaluating the risk of immediate collision.

[0050] As an optional implementation, first, the system receives real-time data from various sensors, including vehicle location information, speed, acceleration, and direction, etc. Then, the system combines these data with the GPS positioning information of the test vehicle to build a dynamic model of the surrounding environment. Using preset danger thresholds such as distance from the test vehicle, relative speed, time-to-collision prediction (TTC), etc., the system can identify vehicles that are too close, have too high relative speed, or have a time-to-collision prediction below the threshold, thereby marking them as potential dangerous targets. This method relies on multi-sensor data input and fusion algorithms, and can provide preliminary prediction of dangerous vehicles, suitable for various application scenarios such as urban roads, highways, etc.

[0051] It is worth noting that by determining the target dangerous vehicle, the vehicle's perception of the surrounding environment can be significantly improved, potential threats can be identified in a timely manner, and warning or intervention decisions can be provided to the driver or autonomous driving system, reducing the risk of traffic accidents. By combining real-time location information with dynamic data of surrounding vehicles, the system can accurately assess the danger level of each target vehicle, thereby improving the safety performance of the vehicle without human intervention.

[0052] Step S106, obtaining vehicle driving information and collision point type of the target dangerous vehicle.

[0053] In the technical solution provided by the above step S106 of the present application, the system first collects real-time data about the target vehicle from various sensors such as radar, camera, lidar, etc., including but not limited to location, speed, acceleration, heading angle, etc. driving parameters. Specifically, the above data is processed by a sensor data fusion algorithm to generate more accurate and comprehensive vehicle driving information.

[0054] Further, using the fused vehicle driving information, the system calculates the relative distance, relative speed, and time-to-collision prediction (TTC) between the target vehicle and the test vehicle. Based on this information, the system can further determine the collision point assignment, and then identify different collision point types such as forward crossing, rear crossing, or side collision through the collision point assignment. Specifically, the coordinates of the collision point are calculated and analyzed, and according to the positive or negative value and the absolute value of the collision point coordinate assignment, the type of the collision point is determined.

[0055] As another optional implementation, through the vehicle-mounted V2X communication technology, the test vehicle can directly receive vehicle driving information from the target dangerous vehicle, including location, speed, heading angle, etc., without relying on local sensor data. The system analyzes the received vehicle driving information and combines its own driving state to analyze the collision probability and determine the collision point type.

[0056] Optionally, the V2X communication technology has the advantage of being able to obtain the dynamic information of the target vehicle in advance, including the future driving intention and path change, so that the collision point type prediction is more accurate and the warning is more timely.

[0057] Further, the system will analyze the results including the driving information of the target vehicle and the collision point type, and share them with each component of the driving assistance system through the vehicle network or other communication means, so as to take appropriate measures in time, such as adjusting the driving route or triggering the warning.

[0058] It is worth noting that by obtaining the vehicle driving information and collision point type of the target dangerous vehicle, the intelligent driving assistance system can conduct a fine evaluation of the potential collision risk, not only improving the accuracy of dangerous vehicle identification, but also distinguishing different types of collision threats, providing a basis for specific risk response measures. Thus, the system's understanding of complex traffic environment and its immediate reaction speed can be enhanced, providing a data basis for subsequent warning and risk avoidance operations.

[0059] Step S108, input the vehicle driving information and collision point type into the target trajectory prediction model to obtain the driving trajectory of the target dangerous vehicle.

[0060] In the technical solution provided by the above step S108 of the present application, the current position, speed, heading angle and other driving information of the target dangerous vehicle, as well as the collision point type calculated by the time to collision (TTC), are arranged in the input format required by the model.

[0061] Further, the filtered trajectory prediction model is used, which is a simple prediction model that assumes the target vehicle will continue to move at the current speed and direction. That is, the above-mentioned driving information and collision point type data are taken as input to start the model calculation process.

[0062] Specifically, as shown in Figure 3 Based on the model assumption, the future position of the target vehicle within a given time interval can be calculated to form a series of predicted coordinate points, depicting a driving trajectory of the vehicle, representing the future driving path of the vehicle.

[0063] Optionally, the above-mentioned trajectory prediction model can be determined by the specific driving environment of the vehicle. Generally, when the vehicle is driving on a highway, a constant speed model can be used; when the vehicle is driving in a city with frequent acceleration and deceleration, a uniform acceleration model can be used.

[0064] As another optional implementation, key features are extracted from the driving information of the target dangerous vehicle, including dynamic parameters such as the speed, acceleration, and steering angle of the vehicle, as well as classification labels based on the collision point type, as input features for the training model.

[0065] Further, a large amount of historical driving data is used in advance, combined with actual collision event records, to train a machine learning model (such as a deep neural network), which aims to learn to predict the future trajectory of the vehicle from the input features. The real-time collected vehicle driving information and collision point type are input, and the trained machine learning model is applied for prediction. The model can take into account the motion trend and possible behavior pattern of the vehicle, output more complex driving trajectory prediction results, and improve the accuracy of prediction.

[0066] It is worth noting that through the above technical steps, the prediction of the future driving trajectory can be realized, so as to predict the potential collision risk in advance, thereby the system can enhance the cognition of the dangerous vehicle motion trend, and provide the driver or automatic driving system with immediate decision basis to avoid or mitigate possible accidents. By predicting the driving trajectory, the system can evaluate the relative motion of the target vehicle and the test vehicle, and prepare for subsequent warning or risk avoidance operations, which significantly improves the collision prevention capability of the intelligent driving assistance system.

[0067] The above steps S102 to S108 can be obtained. In the present application, the position information of the test vehicle is obtained, the target dangerous vehicle is determined according to the position information, the vehicle driving information and the collision point type of the target dangerous vehicle are obtained, and the vehicle driving information and the collision point type are input into the target trajectory prediction model to obtain the driving trajectory of the target dangerous vehicle, thereby solving the technical problem that in the prior art, when a dangerous target is detected during vehicle driving, the motion direction of the dangerous target is not considered, so that the driving trajectory of the dangerous target cannot be accurately predicted.

[0068] The above method of the embodiment will be further described in detail below.

[0069] Step S1041, determining a first dangerous area and a second dangerous area according to the position information, wherein the second dangerous area is larger than the first dangerous area;

[0070] Step S1042, determining a plurality of initial dangerous vehicles according to the first dangerous area and the second dangerous area;

[0071] Step S1043, calculating a plurality of collision durations of the plurality of initial dangerous vehicles, wherein each initial dangerous vehicle corresponds to a collision duration;

[0072] Step S1044, determining a target dangerous vehicle from the plurality of initial dangerous vehicles according to the plurality of collision durations.

[0073] In this embodiment, first, according to the real-time position information of the test vehicle, the system defines two dangerous areas, namely a first dangerous area and a second dangerous area. The first dangerous area is a high-risk area with a short distance from the test vehicle, a high relative speed, or a short predicted collision time, while the second dangerous area is broader and contains vehicles that are slightly farther away but still pose a threat. The second dangerous area is set to have a larger range than the first dangerous area, aiming to capture more potential sources of danger.

[0074] Specifically, within the first and second dangerous areas, the system will automatically screen a plurality of initial dangerous vehicles based on position information and collision time. These vehicles are preliminarily judged to pose a threat to the test vehicle due to their position or motion state.

[0075] Further, for each of the initially screened initial dangerous vehicles, the system calculates the collision time (TTC) between the test vehicle and the initial dangerous vehicle. Optionally, the collision time is an algorithm that predicts the time required for a possible collision between two vehicles by analyzing their relative positions, speeds, and directions. Based on the calculated collision time, the system further analyzes and determines the target dangerous vehicle, i.e., the vehicle with a collision time below a predetermined threshold is identified as a target dangerous vehicle, meaning that it has a risk of colliding with the test vehicle in a short time.

[0076] As an optional implementation, sensors such as radar, cameras, and lidar are used to collect parameters such as the position, speed, and direction of surrounding vehicles. According to the driving environment of the test vehicle, such as urban streets or highways, the range of the first and second dangerous areas is dynamically adjusted, and vehicles within these areas are screened as preliminary dangerous vehicle candidates. For the initially screened vehicles, the sensor data is used to calculate the collision time, and if the collision time is below the set threshold, the vehicle is marked as a target dangerous vehicle for subsequent warning or risk avoidance operations.

[0077] As another optional implementation, through V2X communication technology between vehicles, real-time driving information such as position and speed is received from surrounding vehicles. Based on the received driving information, the system automatically defines the first and second dangerous areas and identifies vehicles within these areas as initial dangerous vehicles. For the initially identified vehicles, the system uses V2X data to calculate the collision time, quickly screening out target dangerous vehicles that will pose a threat to the test vehicle, providing immediate protection measures for driving safety.

[0078] It is worth noting that by defining different dangerous areas based on position information and combining collision time calculation, fast and accurate screening of target dangerous vehicles can be achieved, thereby not only improving the response speed of the intelligent driving assistance system but also enhancing the risk assessment capability of the system, ensuring timely identification of immediate dangers.

[0079] Step S10421, obtaining a plurality of first dangerous vehicles in the first dangerous area;

[0080] Step S10422, obtaining a plurality of second initial dangerous vehicles in the second dangerous area;

[0081] Step S10423, calculating a plurality of dangerous collision durations of the plurality of second initial dangerous vehicles, wherein each second initial dangerous vehicle corresponds to a dangerous collision duration;

[0082] Step S10424, determining a plurality of second dangerous vehicles from the plurality of second initial dangerous vehicles according to the plurality of dangerous collision durations;

[0083] Step S10425, determining a plurality of initial dangerous vehicles according to the plurality of first dangerous vehicles and the plurality of second dangerous vehicles.

[0084] In this embodiment, as shown in FIG. 4, the information of all vehicles in the first dangerous area is obtained, which are usually close to the test vehicle and have high instant threat. Therefore, the definition of this area is based on the real-time position of the test vehicle, aiming to quickly lock the potential danger source at close distance. Figure 2

[0085] Further, the second dangerous area has a wider range, aiming to include those vehicles which are far away but may become a future threat due to high-speed approach or sudden lane change. The system needs to obtain the information of a plurality of vehicles in this area as preliminary dangerous vehicle candidates. For each preliminary selected vehicle in the second dangerous area, the dangerous collision duration between them and the test vehicle is calculated. Next, from the preliminary candidate vehicles in the second dangerous area, according to the calculated dangerous collision duration, those vehicles with collision duration below a preset threshold are selected as high-risk second dangerous vehicles.

[0086] Finally, the plurality of first dangerous vehicles in the first dangerous area and the plurality of selected second dangerous vehicles are integrated to form a complete initial dangerous vehicle set, which constitutes a comprehensive overview of the potential threats to the test vehicle in the near future and the future.

[0087] It is worth mentioning that by dividing the vehicle identification into first and second dangerous areas and screening based on collision duration, the system can effectively determine the plurality of initial dangerous vehicles that pose instant or future potential threats to the test vehicle. This hierarchical vehicle identification and screening strategy not only improves the coverage and accuracy of dangerous vehicle monitoring, but also provides a solid foundation for subsequent warning, obstacle avoidance path planning and other operations, independently improves the safety and response efficiency of the intelligent driving assistance system, and enhances the self-protection ability of the vehicle in complex traffic environment.

[0088] ​Step S201, compare the plurality of dangerous collision time lengths with the preset collision time length, and obtain a first comparison result;

[0089] Step S202, determine a plurality of second dangerous vehicles according to the first comparison result, wherein the second dangerous vehicle is a vehicle whose dangerous collision time length is greater than the preset collision time length.

[0090] In this embodiment, the calculated dangerous collision time length data of the plurality of vehicles in the second region is compared with the system preset collision time length threshold one by one to generate a series of comparison results. Further, based on the first comparison result obtained in the first step, the system can screen out vehicles whose dangerous collision time length is greater than the preset collision time length threshold, and these vehicles are the second dangerous vehicles. Although they are currently safe from the test vehicle, they are expected to pose a threat to the test vehicle at some point in the future due to their motion trend.

[0091] Optionally, the above-mentioned preset collision time length can be determined by those skilled in the art according to the actual working condition, which is not specifically limited here, and in the present application, it can be set to 5 seconds.

[0092] As an optional implementation, different collision time lengths are set with corresponding risk levels, for example, TTC below 3 seconds is considered as extremely high risk, 3 to 5 seconds is considered as high risk, and so on. Then map the TTC value of the second initial dangerous vehicle to the corresponding risk level, and then compare it with the risk level corresponding to the preset collision time length threshold. Keep those vehicles whose risk level mapped by TTC value is higher than or equal to the preset threshold, and classify them as second dangerous vehicles.

[0093] As another optional implementation, real-time driving data and predicted TTC values of surrounding vehicles are continuously received from sensors or V2X communication systems. Then, according to the vehicle speed, road conditions and the safety margin designed by the system, a reasonable collision time length threshold is set. Further, compare the TTC value of each second initial dangerous vehicle with the preset threshold, and keep those vehicles whose TTC exceeds the threshold as second dangerous vehicles. As the vehicle driving condition changes, it is also necessary to recalculate the TTC regularly and update the list of second dangerous vehicles to ensure the real-time effectiveness of the warning system.

[0094] It's worth noting that by comparing multiple dangerous collision durations with preset collision durations, the intelligent driving assistance system can dynamically filter out secondary dangerous vehicles that will not pose a direct collision risk to the test vehicle in the short term but may cause danger in the future. This enhances the system's intelligent perception of the surrounding environment, identifies potential threats in advance, and provides the driver or autonomous driving system with more time and information to make decisions, thereby significantly improving driving safety and reducing the risk of accidents. This technological step not only improves the timeliness of warnings but also optimizes resource allocation, avoiding excessive attention to low-risk vehicles and ensuring that the system gives sufficient attention to truly dangerous vehicles.

[0095] Step S301: Construct a two-dimensional coordinate system based on the location information, wherein the origin of the two-dimensional coordinate system is the center point of the test vehicle;

[0096] Step S302: Determine the initial two-dimensional coordinates of each initial dangerous vehicle in the two-dimensional coordinate system;

[0097] Step S303: Obtain the velocity component during the driving process of each initial dangerous vehicle, wherein the velocity component includes a longitudinal velocity component and a lateral velocity component;

[0098] Step S304: Calculate the collision duration of each initial dangerous vehicle based on the initial two-dimensional coordinates, velocity components, and the vehicle width value of each initial dangerous vehicle.

[0099] In this embodiment, a two-dimensional coordinate system is created based on the real-time location information of the test vehicle, where the center point of the test vehicle is set as the origin. Using sensor data, the precise position of each initial hazardous vehicle within the aforementioned two-dimensional coordinate system is determined, using (…). x, y Coordinates are used to represent the coordinates, where x The axis represents the horizontal position. y The axis represents the longitudinal position.

[0100] Furthermore, real-time driving data for each initial hazardous vehicle is collected, including the longitudinal component of speed (along...). y The axial velocity (i.e., the speed at which the vehicle moves forward or backward) and the lateral component (along the axis). x The velocity along the axis (i.e., the speed at which the vehicle moves laterally). Combining the initial two-dimensional coordinates, the longitudinal and lateral components of the velocity, and the initial width of the dangerous vehicle, a specific physical model or algorithm is used to estimate the duration of the collision.

[0101] Specifically, the formula for calculating the collision duration is as follows:

[0102]

[0103] in, an absolute value of a longitudinal position of the initial dangerous vehicle, an absolute value of a longitudinal component of a speed of the initial dangerous vehicle during driving, w a body width of the initial dangerous vehicle.

[0104] As an optional implementation, sensors such as radar, camera, and lidar capture the position information and speed parameters of surrounding vehicles in real time. The raw data captured by the sensors are converted into two-dimensional coordinates and speed components for collision duration calculation. After each sensor update, the latest data is used to estimate the collision duration immediately, ensuring the timeliness and accuracy of the calculation results. As the vehicle state changes, the calculation results are continuously updated, and the dangerous vehicle list and warning level are dynamically adjusted.

[0105] As another optional implementation, through vehicle-to-environment communication (V2X) technology, the driving state information such as real-time position and speed information sent by surrounding vehicles is received. The V2X data is synchronized with the real-time position information of the test vehicle in the same coordinate system, and the data is processed to extract the required speed component. Further, the future driving intention or path planning information provided by V2X is used to predictively estimate the collision duration to identify potential dangerous situations in advance. In combination with the predicted collision duration, the potential danger is classified and managed, such as marking vehicles with a collision duration less than a preset threshold as high-risk targets.

[0106] It is worth noting that by constructing a two-dimensional coordinate system centered on the test vehicle, collecting and analyzing the initial position and speed component of surrounding vehicles, and calculating the collision duration, the threat of surrounding vehicles can be accurately quantified. This can improve the identification ability of intelligent driving assistance systems for immediate danger and the accuracy of warnings, reducing unnecessary alarms while avoiding missed reports of real threats. By continuously monitoring and updating the collision duration, the system can dynamically adjust the safety alert for surrounding vehicles, providing timely and accurate danger information to the driver or autonomous driving system, effectively improving the safety factor of road travel.

[0107] Step S1061, coordinates of the collision point are assigned according to the initial two-dimensional coordinates and the speed component;

[0108] Step S1062, the coordinate assignment is compared with the first preset value and the second preset value respectively to obtain a second comparison result, wherein the second preset value is greater than the first preset value;

[0109] Step S1063, the collision point type is determined according to the second comparison result, wherein the collision point type includes forward crossing collision, lateral crossing collision, and rear crossing collision.

[0110] In this embodiment, based on the initial position and speed components of the target vehicle in the two-dimensional coordinate system, the system performs a physical model calculation to predict the future possible collision point coordinates. The calculated collision point coordinate values are compared with the first preset value and the second preset value set in advance, so that the collision point position can be classified according to the comparison result. The first preset value and the second preset value generally represent the boundaries in front of and behind the vehicle, and the second preset value is greater than the first preset value to cover the side range of the vehicle.

[0111] Further, according to the comparison result of the coordinate value, the system can judge whether the collision point is in front of, beside or behind the vehicle, so as to determine which type of collision point belongs to: front cross collision, side cross collision or rear cross collision.

[0112] Optionally, the collision point type refers to the relative position of the collision event, which can be divided into front cross collision, side cross collision and rear cross collision, and is used to guide the intelligent driving system to take appropriate measures.

[0113] Specifically, the formula for calculating the collision point coordinates is as follows:

[0114]

[0115] wherein, is the absolute value of the longitudinal position of the initial dangerous vehicle, is the lateral position of the initial dangerous vehicle, is the absolute value of the longitudinal component of the speed of the initial dangerous vehicle during driving, is the absolute value of the lateral component of the speed of the initial dangerous vehicle during driving.

[0116] Optionally, comparing the calculated collision point coordinate values with the first preset value and the second preset value set in advance specifically includes:

[0117] When is greater than 0, it can be determined that the target dangerous vehicle is a front cross collision;

[0118] When is less than 0 and is less than L, it can be determined that the target dangerous vehicle is a side cross collision, wherein L is the length of the body of the target dangerous vehicle;

[0119] When is less than -L, it can be determined that the target dangerous vehicle is a rear cross collision.

[0120] It is worth mentioning that through the above steps, the intelligent driving assistance system can accurately obtain the collision point type, that is, determine whether the collision risk comes from the front, side or rear, thereby enhancing the risk identification capability of the system and making the early warning more refined and targeted. The system can quickly adjust the early warning strategy and evasive action according to the real-time information of the collision point type, thereby significantly improving the safety of driving, reducing the probability of sudden collision accidents, and enhancing the adaptability and safety of intelligent driving vehicles in complex traffic environments.

[0121] Step S110, determining the driving environment of the test vehicle;

[0122] Step S112, determining the target trajectory prediction model from the plurality of trajectory prediction models according to the driving environment.

[0123] In this embodiment, first, the system needs to determine the driving environment in which the test vehicle is located. This environment may include but is not limited to highways, urban streets, rural roads, traffic-intensive areas or open areas. The determination of the driving environment depends on various factors such as vehicle speed, road type, traffic flow and the characteristics of the surrounding environment (such as whether there are intersections, turns or obstacles, etc.).

[0124] Further, after identifying the current driving environment, the system selects the model most suitable for the current environment from the pre-set plurality of trajectory prediction model library as the target trajectory prediction model. Different driving environments may require different models to accurately predict the behavior of surrounding targets, such as uniform speed model for vehicle prediction on smooth traffic highway, and uniform acceleration model for frequent acceleration and deceleration in urban road scenarios.

[0125] Optionally, the trajectory prediction model refers to a mathematical model used to estimate the possible motion trajectory of surrounding targets (such as other vehicles, pedestrians or obstacles) in a certain period of time in the future, which can be based on historical data, current driving state or other external factors for prediction.

[0126] As an optional implementation, real-time data about vehicle speed, acceleration, direction and surrounding target dynamic information is collected by using on-board sensors such as radar, camera and GPS. Based on the sensor data, the system analyzes information such as road type, traffic density and obstacle distribution to determine the specific driving environment. Further, according to the characteristics of the environment, the system automatically selects the matching prediction model. For example, when the system identifies that the test vehicle is driving at a stable speed on the highway, a uniform speed model will be selected; while in a complex urban traffic environment, it may switch to a more complex prediction model, such as a model containing acceleration and deceleration behavior.

[0127] It is worth mentioning that by identifying the specific driving environment of the test vehicle and selecting the most suitable trajectory prediction model accordingly, the system can improve the accuracy and relevance of the prediction, thereby improving the adaptability and prediction performance of the intelligent driving assistance system in different environments. Proper prediction model selection ensures that the system can make more intelligent decisions under certain conditions, effectively avoiding potential traffic accidents and improving driving safety and comfort.

[0128] Through the description of the above embodiments, those skilled in the art can clearly understand that the method according to the above embodiments can be realized by means of software and the necessary general hardware platform, of course, it can also be realized by hardware, but in many cases the former is a better embodiment. Based on such understanding, the technical solutions of the present application or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes a plurality of instructions for making a terminal device (which can be a mobile phone, computer, server, or grid device, etc.) execute the method of each embodiment of the present application.

[0129] In this embodiment, a vehicle trajectory prediction device is also provided, which is used to implement the above embodiments and preferred embodiments, which have been described and will not be repeated. As used below, the term "module" can be a combination of software and / or hardware that implements a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, hardware or a combination of software and hardware is also possible and is contemplated.

[0130] Figure 4 is a structural block diagram of a vehicle trajectory prediction device 400 according to an embodiment of the present application, as shown in Figure 4 The device includes a first acquisition module 401, a first determination module 402, a second acquisition module 403, and an input module 404.

[0131] The first acquisition module 401 is configured to acquire position information of a test vehicle.

[0132] The first determination module 402 is configured to determine a target dangerous vehicle according to the position information.

[0133] The second acquisition module 403 is configured to acquire vehicle driving information and a collision point type of the target dangerous vehicle.

[0134] The input module 404 is configured to input the vehicle driving information and the collision point type into a target trajectory prediction model to obtain a driving trajectory of the target dangerous vehicle.

[0135] Optionally, the first determining module 402 comprises: a first determining unit, configured to determine the first dangerous area and the second dangerous area according to the position information, wherein the second dangerous area is larger than the first dangerous area; a second determining unit, configured to determine a plurality of initial dangerous vehicles according to the first dangerous area and the second dangerous area; a first calculating unit, configured to calculate a plurality of collision durations of the plurality of initial dangerous vehicles, wherein each initial dangerous vehicle corresponds to one collision duration; and a third determining unit, configured to determine a target dangerous vehicle from the plurality of initial dangerous vehicles according to the plurality of collision durations.

[0136] Optionally, the second determining unit comprises: a first obtaining sub-unit, configured to obtain a plurality of first dangerous vehicles in the first dangerous area; a second obtaining sub-unit, configured to obtain a plurality of second initial dangerous vehicles in the second dangerous area; a first calculating sub-unit, configured to calculate a plurality of dangerous collision durations of the plurality of second initial dangerous vehicles, wherein each second initial dangerous vehicle corresponds to one dangerous collision duration; a first determining sub-unit, configured to determine a plurality of second dangerous vehicles from the plurality of second initial dangerous vehicles according to the plurality of dangerous collision durations; and a second determining sub-unit, configured to determine the plurality of initial dangerous vehicles according to the plurality of first dangerous vehicles and the plurality of second dangerous vehicles.

[0137] Optionally, the first determining sub-unit comprises: a comparing sub-unit, configured to compare the plurality of dangerous collision durations with a preset collision duration to obtain a first comparison result; and a third determining sub-unit, configured to determine the plurality of second dangerous vehicles according to the first comparison result, wherein the second dangerous vehicle is a vehicle with a dangerous collision duration greater than the preset collision duration.

[0138] Optionally, the first calculating unit comprises: a constructing sub-unit, configured to construct a two-dimensional coordinate system according to the position information, wherein the origin of the two-dimensional coordinate system is the center point of the test vehicle; a fourth determining sub-unit, configured to determine an initial two-dimensional coordinate of each initial dangerous vehicle in the two-dimensional coordinate system; a third obtaining sub-unit, configured to obtain a speed component of each initial dangerous vehicle in a driving process, wherein the speed component comprises a speed longitudinal component and a speed transverse component; and a second calculating sub-unit, configured to calculate the collision duration of each initial dangerous vehicle according to the initial two-dimensional coordinate, the speed component, and a vehicle body width value of each initial dangerous vehicle.

[0139] Optionally, the second obtaining module 403 comprises: a second calculating unit, configured to calculate a coordinate assignment of the collision point according to the initial two-dimensional coordinate and the speed component; a comparing unit, configured to compare the coordinate assignment with a first preset value and a second preset value respectively to obtain a second comparison result, wherein the second preset value is greater than the first preset value; and a fourth determining unit, configured to determine the collision point type according to the second comparison result, wherein the collision point type comprises a front cross collision, a side cross collision, and a rear cross collision.

[0140] Optionally, the vehicle trajectory prediction device 400 further comprises a second determining module configured to determine a driving environment of the test vehicle; and a third determining module configured to determine the target trajectory prediction model from the plurality of trajectory prediction models according to the driving environment.

[0141] Embodiments of the present application also provide a vehicle comprising a memory and a processor, wherein the memory stores a computer program, and the processor is configured to execute the computer program to perform the vehicle trajectory prediction method described above.

[0142] Optionally, in the present embodiment, the vehicle described above can be configured to store a computer program for performing the following steps:

[0143] Step S102, obtaining position information of the test vehicle;

[0144] Step S104, determining a target dangerous vehicle according to the position information;

[0145] Step S106, obtaining vehicle driving information and a collision point type of the target dangerous vehicle;

[0146] Step S108, inputting the vehicle driving information and the collision point type into the target trajectory prediction model to obtain a driving trajectory of the target dangerous vehicle.

[0147] Optionally, when the processor executes the program, the following steps are further implemented: determining a first dangerous area and a second dangerous area according to the position information, wherein the second dangerous area is larger than the first dangerous area; determining a plurality of initial dangerous vehicles according to the first dangerous area and the second dangerous area; calculating a plurality of collision durations of the plurality of initial dangerous vehicles, wherein each initial dangerous vehicle corresponds to one collision duration; and determining the target dangerous vehicle from the plurality of initial dangerous vehicles according to the plurality of collision durations.

[0148] Optionally, when the processor executes the program, the following steps are further implemented: obtaining a plurality of first initial dangerous vehicles within the first dangerous area; obtaining a plurality of second initial dangerous vehicles within the second dangerous area; calculating a plurality of dangerous collision durations of the plurality of second initial dangerous vehicles, wherein each second initial dangerous vehicle corresponds to one dangerous collision duration; determining a plurality of second dangerous vehicles from the plurality of second initial dangerous vehicles according to the plurality of dangerous collision durations; and determining the plurality of initial dangerous vehicles according to the plurality of first dangerous vehicles and the plurality of second dangerous vehicles.

[0149] Optionally, when the processor executes the program, the following steps are further implemented: comparing the plurality of dangerous collision durations with a preset collision duration to obtain a first comparison result; and determining the plurality of second dangerous vehicles according to the first comparison result, wherein the second dangerous vehicle is a vehicle with a dangerous collision duration greater than the preset collision duration.

[0150] Optionally, the processor, when executing the program, further implements the following steps: constructing a two-dimensional coordinate system according to the position information, wherein the origin of the two-dimensional coordinate system is the center point of the test vehicle; determining an initial two-dimensional coordinate of each initial dangerous vehicle in the two-dimensional coordinate system; obtaining a speed component of each initial dangerous vehicle in the driving process, wherein the speed component includes a speed longitudinal component and a speed transverse component; and calculating a collision duration of each initial dangerous vehicle according to the initial two-dimensional coordinate, the speed component, and a vehicle body width value of each initial dangerous vehicle.

[0151] Optionally, the processor, when executing the program, further implements the following steps: calculating a coordinate assignment of the collision point according to the initial two-dimensional coordinate and the speed component; comparing the coordinate assignment with a first preset value and a second preset value respectively to obtain a second comparison result, wherein the second preset value is greater than the first preset value; and determining a collision point type according to the second comparison result, wherein the collision point type includes a front cross collision, a side cross collision, and a rear cross collision.

[0152] Optionally, the processor, when executing the program, further implements the following steps: determining a driving environment of the test vehicle; and determining a target trajectory prediction model from a plurality of trajectory prediction models according to the driving environment.

[0153] Optionally, specific examples in the embodiment can refer to the examples described in the above embodiments and optional implementation manners, and the embodiment will not be described here again.

[0154] Embodiments of the application also provide an electronic device including a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to execute the above-mentioned vehicle trajectory prediction method.

[0155] Optionally, in the embodiment, the electronic device can be configured to store a computer program for executing the following steps:

[0156] Step S102: obtaining position information of a test vehicle;

[0157] Step S104: determining a target dangerous vehicle according to the position information;

[0158] Step S106: obtaining vehicle driving information and a collision point type of the target dangerous vehicle;

[0159] Step S108: inputting the vehicle driving information and the collision point type into the target trajectory prediction model to obtain a driving trajectory of the target dangerous vehicle.

[0160] Optionally, the processor, when executing the program, further implements the following steps: determining the first dangerous area and the second dangerous area according to the position information, wherein the second dangerous area is larger than the first dangerous area; determining a plurality of initial dangerous vehicles according to the first dangerous area and the second dangerous area; calculating a plurality of collision durations of the plurality of initial dangerous vehicles, wherein each initial dangerous vehicle corresponds to a collision duration; and determining a target dangerous vehicle from the plurality of initial dangerous vehicles according to the plurality of collision durations.

[0161] Optionally, the processor, when executing the program, further implements the following steps: obtaining a plurality of first initial dangerous vehicles in the first dangerous area; obtaining a plurality of second initial dangerous vehicles in the second dangerous area; calculating a plurality of dangerous collision durations of the plurality of second initial dangerous vehicles, wherein each second initial dangerous vehicle corresponds to a dangerous collision duration; determining a plurality of second dangerous vehicles from the plurality of second initial dangerous vehicles according to the plurality of dangerous collision durations; and determining the plurality of initial dangerous vehicles according to the plurality of first initial dangerous vehicles and the plurality of second dangerous vehicles.

[0162] Optionally, the processor, when executing the program, further implements the following steps: comparing the plurality of dangerous collision durations with a preset collision duration to obtain a first comparison result; and determining the plurality of second dangerous vehicles according to the first comparison result, wherein a second dangerous vehicle is a vehicle with a dangerous collision duration greater than the preset collision duration.

[0163] Optionally, the processor, when executing the program, further implements the following steps: constructing a two-dimensional coordinate system according to the position information, wherein the origin of the two-dimensional coordinate system is the center point of the test vehicle; determining an initial two-dimensional coordinate of each initial dangerous vehicle in the two-dimensional coordinate system; obtaining a speed component of each initial dangerous vehicle during driving, wherein the speed component includes a speed longitudinal component and a speed transverse component; and calculating the collision duration of each initial dangerous vehicle according to the initial two-dimensional coordinate, the speed component, and a vehicle body width value of each initial dangerous vehicle.

[0164] Optionally, the processor, when executing the program, further implements the following steps: calculating a coordinate assignment of a collision point according to the initial two-dimensional coordinate and the speed component; comparing the coordinate assignment with a first preset value and a second preset value respectively to obtain a second comparison result, wherein the second preset value is greater than the first preset value; and determining a collision point type according to the second comparison result, wherein the collision point type includes a front cross collision, a side cross collision, and a rear cross collision.

[0165] Optionally, the processor, when executing the program, further implements the following steps: determining a driving environment of the test vehicle; and determining a target trajectory prediction model from a plurality of trajectory prediction models according to the driving environment.

[0166] Optionally, specific examples in the present embodiment can refer to the examples described in the above embodiments and optional implementation manners, which will not be described herein again.

[0167] The embodiment of the present application also provides a computer readable storage medium, wherein the computer readable storage medium stores a computer program, and the computer program is configured to execute the above-mentioned vehicle trajectory prediction method when running on a computer or a processor.

[0168] Optionally, in the embodiment, the computer readable storage medium is configured to store a computer program for executing the following steps:

[0169] S102, acquiring position information of a test vehicle;

[0170] S104, determining a target dangerous vehicle according to the position information;

[0171] S106, acquiring vehicle driving information and a collision point type of the target dangerous vehicle;

[0172] S108, inputting the vehicle driving information and the collision point type into a target trajectory prediction model to obtain a driving trajectory of the target dangerous vehicle.

[0173] Optionally, the storage medium is configured to store program code for executing the following steps: determining a first dangerous area and a second dangerous area according to the position information, wherein the second dangerous area is larger than the first dangerous area; determining a plurality of initial dangerous vehicles according to the first dangerous area and the second dangerous area; calculating a plurality of collision durations of the plurality of initial dangerous vehicles, wherein each initial dangerous vehicle corresponds to one collision duration; and determining the target dangerous vehicle from the plurality of initial dangerous vehicles according to the plurality of collision durations.

[0174] Optionally, the storage medium is configured to store program code for executing the following steps: acquiring a plurality of first initial dangerous vehicles in the first dangerous area; acquiring a plurality of second initial dangerous vehicles in the second dangerous area; calculating a plurality of dangerous collision durations of the plurality of second initial dangerous vehicles, wherein each second initial dangerous vehicle corresponds to one dangerous collision duration; determining a plurality of second dangerous vehicles from the plurality of second initial dangerous vehicles according to the plurality of dangerous collision durations; and determining the plurality of initial dangerous vehicles according to the plurality of first initial dangerous vehicles and the plurality of second dangerous vehicles.

[0175] Optionally, the storage medium is configured to store program code for executing the following steps: comparing the plurality of dangerous collision durations with a preset collision duration to obtain a first comparison result; and determining the plurality of second dangerous vehicles according to the first comparison result, wherein the second dangerous vehicle is a vehicle with a dangerous collision duration greater than the preset collision duration.

[0176] Optionally, the storage medium is configured to store program code for performing the following steps: constructing a two-dimensional coordinate system according to the position information, wherein the origin of the two-dimensional coordinate system is the center point of the test vehicle; determining an initial two-dimensional coordinate of each initial dangerous vehicle in the two-dimensional coordinate system; obtaining a speed component of each initial dangerous vehicle during driving, wherein the speed component includes a speed longitudinal component and a speed transverse component; and calculating a collision duration of each initial dangerous vehicle according to the initial two-dimensional coordinate, the speed component, and a vehicle body width value of each initial dangerous vehicle.

[0177] Optionally, the storage medium is configured to store program code for performing the following steps: calculating a coordinate assignment of the collision point according to the initial two-dimensional coordinate and the speed component; comparing the coordinate assignment with a first preset value and a second preset value respectively to obtain a second comparison result, wherein the second preset value is greater than the first preset value; and determining a collision point type according to the second comparison result, wherein the collision point type includes a front cross collision, a side cross collision, and a rear cross collision.

[0178] Optionally, the storage medium is configured to store program code for performing the following steps: determining a driving environment of the test vehicle; and determining a target trajectory prediction model from a plurality of trajectory prediction models according to the driving environment.

[0179] Optionally, specific examples in the embodiment can refer to the examples described in the above embodiments and optional implementation manners, and the embodiment will not be described here again.

[0180] The embodiment of the application further provides a computer program product comprising a computer program, wherein the computer program implements the steps of the vehicle trajectory prediction method described above when executed by a processor.

[0181] Optionally, in the embodiment, the computer program product can be configured to store a computer program for performing the following steps:

[0182] Step S102: obtaining position information of a test vehicle;

[0183] Step S104: determining a target dangerous vehicle according to the position information;

[0184] Step S106: obtaining vehicle driving information and a collision point type of the target dangerous vehicle;

[0185] Step S108: inputting the vehicle driving information and the collision point type into the target trajectory prediction model to obtain a driving trajectory of the target dangerous vehicle.

[0186] Optionally, the computer program further implements the following steps when executed by the computer: determining the first dangerous area and the second dangerous area according to the position information, wherein the second dangerous area is larger than the first dangerous area; determining a plurality of initial dangerous vehicles according to the first dangerous area and the second dangerous area; calculating a plurality of collision durations of the plurality of initial dangerous vehicles, wherein each initial dangerous vehicle corresponds to a collision duration; and determining a target dangerous vehicle from the plurality of initial dangerous vehicles according to the plurality of collision durations.

[0187] Optionally, the computer program further implements the following steps when executed by the computer: obtaining a plurality of first initial dangerous vehicles in the first dangerous area; obtaining a plurality of second initial dangerous vehicles in the second dangerous area; calculating a plurality of dangerous collision durations of the plurality of second initial dangerous vehicles, wherein each second initial dangerous vehicle corresponds to a dangerous collision duration; determining a plurality of second dangerous vehicles from the plurality of second initial dangerous vehicles according to the plurality of dangerous collision durations; and determining the plurality of initial dangerous vehicles according to the plurality of first initial dangerous vehicles and the plurality of second dangerous vehicles.

[0188] Optionally, the computer program further implements the following steps when executed by the computer: comparing the plurality of dangerous collision durations with a preset collision duration to obtain a first comparison result; and determining the plurality of second dangerous vehicles according to the first comparison result, wherein a second dangerous vehicle is a vehicle with a dangerous collision duration greater than the preset collision duration.

[0189] Optionally, the computer program further implements the following steps when executed by the computer: constructing a two-dimensional coordinate system according to the position information, wherein the origin of the two-dimensional coordinate system is the center point of the test vehicle; determining an initial two-dimensional coordinate of each initial dangerous vehicle in the two-dimensional coordinate system; obtaining a speed component of each initial dangerous vehicle during driving, wherein the speed component includes a speed longitudinal component and a speed transverse component; and calculating the collision duration of each initial dangerous vehicle according to the initial two-dimensional coordinate, the speed component, and a vehicle body width value of each initial dangerous vehicle.

[0190] Optionally, the computer program further implements the following steps when executed by the computer: calculating a coordinate assignment of a collision point according to the initial two-dimensional coordinate and the speed component; comparing the coordinate assignment with a first preset value and a second preset value respectively to obtain a second comparison result, wherein the second preset value is greater than the first preset value; and determining a collision point type according to the second comparison result, wherein the collision point type includes a front cross collision, a side cross collision, and a rear cross collision.

[0191] Optionally, the computer program further implements the following steps when executed by the computer: determining a driving environment of the test vehicle; and determining a target trajectory prediction model from a plurality of trajectory prediction models according to the driving environment.

[0192] Optionally, specific examples in the present embodiment can refer to the examples described in the above embodiments and optional implementation manners, which will not be described herein again.

[0193] In the above-described embodiments of the present application, the description of each embodiment focuses on different aspects, and the parts not described in detail in a certain embodiment can be referred to the relevant description of other embodiments.

[0194] In some embodiments provided in the present application, it should be understood that the disclosed technology can be implemented in other ways. The above-described device embodiments are only illustrative, for example, the division of units can be a logical function division, and actual implementation can have another division manner, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the units or modules shown or discussed can be indirect coupling or communication connection through some interfaces, units or modules, which can be electrical or other forms.

[0195] The units described as separate components can or can not be physically separate, and the components shown as units can or can not be physical units, that is, they can be located in one place or distributed on multiple units. Part or all of the units can be selected according to actual needs to achieve the purpose of the present embodiment.

[0196] In addition, each functional unit in each embodiment of the present application can be integrated in one processing unit, or each unit can exist physically, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of a software functional unit.

[0197] If the integrated unit is realized in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application essentially or the part that contributes to the prior art or the whole or part of the technical solutions can be embodied in the form of a software product, which is stored in a storage medium and includes a plurality of instructions for making a computer device (which can be a personal computer, a server or a grid device, etc.) execute all or part of the steps of the embodiments of the present application. The foregoing storage medium includes: a U disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a mobile hard disk, a magnetic disk or an optical disk, and various program code storage media.

[0198] The above merely is the preferred embodiment of the present application, it should be pointed out that, for ordinary skilled in the art, without departing from the principles of the present application, can also make a number of improvements and refinements, these improvements and refinements should also be considered as the protection scope of the present application.

Claims

1. A method of predicting a trajectory of a vehicle, characterized by, The method comprises: obtaining position information of a test vehicle; determining a target dangerous vehicle according to the position information; obtaining vehicle driving information and a collision point type of the target dangerous vehicle; inputting the vehicle driving information and the collision point type into a target trajectory prediction model to obtain a driving trajectory of the target dangerous vehicle.

2. The method of predicting a trajectory of a vehicle according to claim 1, wherein, Determining the target dangerous vehicle according to the position information comprises: determining a first dangerous area and a second dangerous area according to the position information, wherein the second dangerous area is larger than the first dangerous area; determining a plurality of initial dangerous vehicles according to the first dangerous area and the second dangerous area; calculating a plurality of collision durations of the plurality of initial dangerous vehicles, wherein each initial dangerous vehicle corresponds to a collision duration; determining the target dangerous vehicle from the plurality of initial dangerous vehicles according to the plurality of collision durations.

3. The method of predicting a trajectory of a vehicle according to claim 2, wherein, Determining the plurality of initial dangerous vehicles according to the first dangerous area and the second dangerous area comprises: obtaining a plurality of first dangerous vehicles in the first dangerous area; obtaining a plurality of second initial dangerous vehicles in the second dangerous area; calculating a plurality of dangerous collision durations of the plurality of second initial dangerous vehicles, wherein each second initial dangerous vehicle corresponds to a dangerous collision duration; determining a plurality of second dangerous vehicles from the plurality of second initial dangerous vehicles according to the plurality of dangerous collision durations; determining the plurality of initial dangerous vehicles according to the plurality of first dangerous vehicles and the plurality of second dangerous vehicles.

4. The method of predicting a trajectory of a vehicle according to claim 3, wherein, Determining the plurality of second dangerous vehicles from the plurality of second initial dangerous vehicles according to the plurality of dangerous collision durations comprises: comparing the plurality of dangerous collision durations with a preset collision duration to obtain a first comparison result; determining the plurality of second dangerous vehicles according to the first comparison result, wherein the second dangerous vehicle is a vehicle with a dangerous collision duration greater than the preset collision duration.

5. The method of predicting a trajectory of a vehicle according to claim 2, wherein, Calculating the plurality of collision durations of the plurality of initial dangerous vehicles comprises: constructing a two-dimensional coordinate system according to the position information, wherein the origin of the two-dimensional coordinate system is the center point of the test vehicle; determining an initial two-dimensional coordinate of each initial dangerous vehicle in the two-dimensional coordinate system; obtaining a speed component during driving of each initial dangerous vehicle, wherein the speed component comprises a speed longitudinal component and a speed transverse component; calculating a collision duration of each initial dangerous vehicle according to the initial two-dimensional coordinate, the speed component, and a vehicle body width value of each initial dangerous vehicle.

6. The method of predicting a trajectory of a vehicle according to claim 5, wherein, Obtaining the collision point type comprises: calculating a coordinate assignment of a collision point according to the initial two-dimensional coordinate and the speed component; comparing the coordinate assignment with a first preset value and a second preset value respectively to obtain a second comparison result, wherein the second preset value is greater than the first preset value; determining the collision point type according to the second comparison result, wherein the collision point type comprises a forward crossing collision, a lateral crossing collision, and a rearward crossing collision.

7. The method of predicting a trajectory of a vehicle according to claim 1, wherein, The method further comprises: determining a driving environment of the test vehicle; determining the target trajectory prediction model from a plurality of trajectory prediction models according to the driving environment.

8. A device for predicting a trajectory of a vehicle, characterized in that The method comprises: A first obtaining module is configured to obtain position information of a test vehicle; A first determining module is configured to determine a target dangerous vehicle according to the position information; A second obtaining module is configured to obtain vehicle driving information and a collision point type of the target dangerous vehicle; An input module is configured to input the vehicle driving information and the collision point type into a target trajectory prediction model to obtain a driving trajectory of the target dangerous vehicle.

9. A vehicle comprising a memory and a processor, characterized in that The memory stores a computer program, and the processor is configured to run the computer program to perform the vehicle trajectory prediction method in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer readable storage medium stores a computer program, and the computer program is configured to perform the vehicle trajectory prediction method in any one of claims 1 to 7 when running on a computer or a processor.

Images