Collision risk assessment method and apparatus, electronic device, and storage medium

Abstract

The present application provides a collision risk assessment method, comprising: acquiring a trajectory to be assessed of an ego vehicle and a predicted trajectory of another vehicle within a preset range; transforming the trajectory to be assessed and the predicted trajectory into a same S-T coordinate system, wherein S represents the traveling distance of a vehicle along a current lane in which the ego vehicle travels, and T represents time; and on the basis of the trajectory to be assessed and the predicted trajectory in the S-T coordinate system, determining whether there is a collision risk for the ego vehicle. In embodiments of the present application, a trajectory to be assessed of an ego vehicle and a predicted trajectory of another vehicle within a preset range are transformed into a same S-T coordinate system, and whether there is a collision risk for the ego vehicle is determined on the basis of the trajectory to be assessed and the predicted trajectory in the S-T coordinate system. In the embodiments of the present application, the S-T coordinate system represents a correspondence between the traveling distance of a vehicle along a current lane and time, and the trajectory of the vehicle in the S-T coordinate system appears in the form of a rectangle, thereby resolving the trade-off between detection density and detection accuracy.

Description

Collision risk assessment methods, devices, electronic equipment and storage media

[0001] This application claims priority to Chinese patent application filed on January 21, 2025, with application number 202510096618.0 and entitled "Collision Risk Assessment Method, Apparatus, Electronic Device and Storage Medium", the entire contents of which are incorporated herein by reference. Technical Field

[0002] This application relates to the field of driver assistance technology, specifically to a collision risk assessment method, device, electronic device, and storage medium. Background Technology

[0003] With the continuous development of automotive technology, more and more cars are equipped with intelligent driving functions. Driving safety is the primary consideration for intelligent driving functions. Driving safety includes not only avoiding collisions with other vehicles and obstacles while driving, but also avoiding collisions with other vehicles. Therefore, collision risk assessment has become of paramount importance.

[0004] In related technologies, the predicted trajectory of the obstacle and the planned trajectory of the vehicle are acquired at preset time intervals. At the same time, it is determined whether there is an intersection point between the predicted trajectory of the obstacle and the planned trajectory of the vehicle in the XY plane projection, thereby determining whether there is a collision risk.

[0005] However, the size of the time interval in the relevant technology directly affects the accuracy of collision risk detection. Too large a time interval will lead to missed detections, while too small a time interval will lead to excessive computational load.

[0006] It should be noted that the information disclosed in the background section of this application is intended only to enhance the understanding of the general background of this application, and should not be construed as an admission or in any way implying that the information constitutes prior art known to those skilled in the art. Summary of the Invention

[0007] In view of this, this application provides a collision risk assessment method, apparatus, electronic device, and storage medium to address the problem in related technologies that time intervals directly affect the accuracy of collision risk detection, with excessively large time intervals leading to missed detections and excessively small time intervals leading to excessive computational load.

[0008] In a first aspect, embodiments of this application provide a collision risk assessment method, including:

[0009] Acquire the trajectory of the vehicle to be evaluated and the predicted trajectories of other vehicles within a preset range;

[0010] The trajectory to be evaluated and the predicted trajectory are transformed to the same ST coordinate system, where S is the distance traveled by the vehicle in the current lane and T is the time.

[0011] Based on the trajectory to be evaluated and the predicted trajectory in the ST coordinate system, it is determined whether the vehicle has a collision risk.

[0012] In this embodiment, the trajectory to be evaluated of the vehicle and the predicted trajectories of other vehicles within a preset range are transformed into the same ST coordinate system. Based on the trajectory to be evaluated and the predicted trajectories in the ST coordinate system, it is determined whether the vehicle has a collision risk. In this embodiment, the ST coordinate system represents the correspondence between the vehicle's travel distance and time in the current lane, and the vehicle's trajectory in the ST coordinate system appears in the form of a rectangle, thus resolving the contradiction between detection density and detection accuracy.

[0013] In one possible implementation, determining whether the vehicle has a collision risk based on the trajectory to be evaluated and the predicted trajectory in the ST coordinate system includes:

[0014] Determine whether the collision detection unit of the trajectory to be evaluated and the collision detection unit of the predicted trajectory coincide in the same time period in the ST coordinate system, wherein the collision detection unit is the trajectory within a preset time period;

[0015] If the collision detection unit of the trajectory to be evaluated and the collision detection unit of the predicted trajectory overlap in the same time period in the ST coordinate system, then it is determined that the vehicle has a collision risk.

[0016] If the collision detection unit of the trajectory to be evaluated and the collision detection unit of the predicted trajectory do not overlap in the same time period in the ST coordinate system, then it is determined that the vehicle has no collision risk.

[0017] In this embodiment, the presence of a collision risk is determined by whether the collision detection units of the trajectory to be evaluated and the collision detection units of the predicted trajectory overlap within the same time period in the ST coordinate system. It can be understood that the collision detection unit represents the trajectory within a preset time period; if the collision detection units overlap, a collision risk is determined.

[0018] In one possible implementation, the preset duration is the minimum step update duration. Before determining whether the collision detection unit of the trajectory to be evaluated in the ST coordinate system coincides with the collision detection unit of the predicted trajectory, the method further includes:

[0019] The minimum step update time is calculated based on the preset decision cycle and the maximum number of decision updates, where the maximum number of decision updates is the maximum number of times the decision is changed within the preset decision cycle.

[0020] In this embodiment, the preset duration is the minimum step update duration. It is understood that the minimum step update duration ensures that every decision step of the vehicle is recorded, and that every decision made by the vehicle can be accurately recorded.

[0021] In one possible implementation, the method further includes:

[0022] The driving style is determined based on the number of decision updates, where the number of decision updates is the number of times the decision is changed within the preset decision period;

[0023] Adjust driving decisions based on the described driving style.

[0024] In this embodiment, driving style is determined based on the number of decision updates, and driving decisions are adjusted accordingly. It is understood that determining driving style helps plan multiple actions within the decision-making cycle, thereby enabling more flexible and efficient vehicle control and avoiding overly conservative driving.

[0025] In one possible implementation, determining the driving style based on the number of decision updates includes:

[0026] If the number of decision updates is greater than or equal to the first preset number, then the driving style is determined to be aggressive.

[0027] If the number of decision updates is less than or equal to the second preset number, then the driving style is determined to be conservative.

[0028] The first preset number of times is greater than the second preset number of times.

[0029] In this embodiment, if the number of decision updates is greater than or equal to a first preset number, the driving style is determined to be aggressive; if the number of decision updates is less than or equal to a second preset number, the driving style is determined to be conservative. It can be understood that if decision updates are too frequent, the driving style is more aggressive; if decision updates are too slow, the driving style is more conservative.

[0030] In one possible implementation, determining whether the collision detection unit of the trajectory to be evaluated in the ST coordinate system coincides with the collision detection unit of the predicted trajectory includes:

[0031] If the distance between the predicted trajectory and the trajectory to be evaluated is less than or equal to a preset distance threshold at the same time, it is determined whether the collision detection unit of the trajectory to be evaluated and the collision detection unit of the predicted trajectory in the ST coordinate system overlap.

[0032] In this embodiment, the collision detection units in the ST coordinate system are only considered to overlap if the distance between the predicted trajectory and the trajectory to be evaluated at the same moment is less than or equal to a preset distance threshold. It is understood that if the distance between the predicted trajectory and the trajectory to be evaluated at the same moment is greater than the preset distance threshold, then there is no risk of collision between the vehicle and other vehicles. Therefore, determining the distance in advance can avoid some unnecessary computational load.

[0033] In one possible implementation, the method further includes:

[0034] Determine whether the distance between the predicted trajectory and the trajectory to be evaluated at the same moment is less than or equal to a preset distance threshold;

[0035] If the distance between the predicted trajectory and the trajectory to be evaluated is greater than a preset distance threshold at the same time, it is determined that the vehicle does not have a collision risk.

[0036] In this embodiment of the application, if the distance between the predicted trajectory and the trajectory to be evaluated at the same time is greater than a preset distance threshold, it can be directly determined that the vehicle does not have a collision risk, thus avoiding some unnecessary computational load.

[0037] Secondly, embodiments of this application provide a collision risk assessment device, including:

[0038] The trajectory acquisition module is used to acquire the trajectory to be evaluated of the vehicle and the predicted trajectories of other vehicles within a preset range;

[0039] The coordinate transformation module is used to transform the trajectory to be evaluated and the predicted trajectory to the same ST coordinate system, where S is the distance traveled by the vehicle in the current lane and T is the time.

[0040] The collision risk determination module is used to determine whether the vehicle has a collision risk based on the trajectory to be evaluated and the predicted trajectory in the ST coordinate system.

[0041] Thirdly, embodiments of this application provide an electronic device, including:

[0042] processor;

[0043] Memory;

[0044] And a computer program, wherein the computer program is stored in the memory, the computer program including instructions that, when executed by the processor, cause the electronic device to perform the method described in any one of the first aspects.

[0045] Fourthly, embodiments of this application provide a computer-readable storage medium, characterized in that the computer-readable storage medium includes a stored program, wherein, when the program is executed, it controls the device where the computer-readable storage medium is located to perform the method described in any one of the first aspects.

[0046] It is understood that the collision risk assessment device provided in the second aspect, the electronic device provided in the third aspect, and the computer-readable storage medium provided in the fourth aspect are used to execute the method provided in this application. Therefore, the beneficial effects they can achieve can be referred to the beneficial effects in the corresponding methods, and will not be repeated here. Attached Figure Description

[0047] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0048] Figure 1 is a schematic diagram of a grid map in the first type of related technology;

[0049] Figure 2 is a schematic diagram illustrating the correspondence between detection density and detection results provided in an embodiment of this application;

[0050] Figure 3 is a flowchart illustrating a collision risk assessment method provided in an embodiment of this application;

[0051] Figure 4 is a schematic diagram of the trajectory to be evaluated and the predicted trajectory in the ST coordinate system provided in an embodiment of this application;

[0052] Figure 5 is a schematic diagram of the trajectory to be evaluated and the predicted trajectory in another ST coordinate system provided in the embodiments of this application;

[0053] Figure 6 is a schematic diagram of the trajectory to be evaluated and the predicted trajectory in another ST coordinate system provided in the embodiments of this application;

[0054] Figure 7 is a schematic diagram of the trajectory to be evaluated and the predicted trajectory in another ST coordinate system provided in the embodiments of this application;

[0055] Figure 8 is a schematic diagram of a collision risk assessment device provided in an embodiment of this application;

[0056] Figure 9 is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0057] To better understand the technical solution of this application, the embodiments of this application will be described in detail below with reference to the accompanying drawings.

[0058] It should be understood that the described embodiments are merely some, not all, of the embodiments in this application. All other embodiments obtained by those skilled in the art based on the embodiments in this application without inventive effort are within the scope of protection of this application.

[0059] The terminology used in the embodiments of this application is for the purpose of describing particular embodiments only and is not intended to be limiting of this application. The singular forms “a,” “the,” and “the” used in the embodiments of this application and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise.

[0060] It should be understood that the term "and / or" used in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this article generally indicates that the preceding and following related objects have an "or" relationship.

[0061] With the continuous development of automotive technology, more and more cars are equipped with intelligent driving functions. Driving safety is the primary consideration for intelligent driving functions. Driving safety includes not only avoiding collisions with other vehicles and obstacles while driving, but also avoiding collisions with other vehicles. Therefore, collision risk assessment has become of paramount importance.

[0062] In the first category of related technologies, the vehicle's trajectory is based on a two-dimensional occupancy grid map. The autonomous vehicle is aware of the grid map showing obstacles at the current moment, and then determines the collision risk of the planned trajectory by sampling and querying whether obstacles occupy it. Furthermore, to simplify the calculation of the grid area covered by the vehicle along the planned trajectory, the vehicle's outline is generally abstracted into simple, regular geometric objects such as circles or rectangles. For ease of understanding, this application also provides a schematic diagram of a grid map in the related technology.

[0063] Referring to Figure 1, which is a schematic diagram of a grid map in the first category of related technologies. As shown in Figure 1, the dashed lines in the grid map represent the planned trajectory of the vehicle, and the vehicle outline in the grid map of Figure 1 is a rectangular outline. In practical applications, the vehicle outline can also be abstracted into other geometric objects, such as a circular outline and a tri-circular outline. When an obstacle exists, the corresponding position in the grid map will be marked with the obstacle's location. By determining whether the planned trajectory of the vehicle coincides with the obstacle, it can be determined whether a collision will occur between the vehicle and the obstacle.

[0064] However, the above method only considers the obstacle occupancy status at the current time and does not take into account the future movement trend of the obstacles, leading to short-sightedness in autonomous vehicles. For example, at the current moment, there may be no obstacles occupying the latter half of the vehicle's planned trajectory, but this does not mean that there will still be no obstacles when the vehicle moves to the latter half of the planned trajectory.

[0065] To address this issue, Type II related technologies typically predict the future trajectory of obstacles and incorporate it into collision risk detection. Compared to the methods provided by Type I related technologies, Type II technologies add a time dimension, thus dynamically considering the obstacle's movement trend. By checking whether the predicted trajectory of the obstacle and the planned trajectory of the vehicle intersect in the XY plane at the same time, it is determined whether a collision will occur between the vehicle and the obstacle. If an intersection exists, it indicates a collision risk.

[0066] In the second category of related technologies, it is necessary to acquire the predicted trajectory of the obstacle and the trajectory of the vehicle at preset time intervals. However, the size of the time interval directly affects the accuracy of collision risk detection. Too large a time interval (i.e., too low a detection density) will lead to missed detections, while too small a time interval (i.e., too high a detection density) will lead to excessive computational load.

[0067] [Corrected according to Rule 91, 13.01.2026] Refer to Figure 2, which is a schematic diagram of the correspondence between detection density and detection results provided in an embodiment of this application. As shown in Figure 2, vehicle C1 is the neighboring vehicle (i.e., the obstacle), vehicle C2 is the vehicle itself, the line corresponding to vehicle C1 is the predicted trajectory of the neighboring vehicle, and the line corresponding to vehicle C2 is the candidate set of the planned trajectory of the vehicle itself. The intersection point P(x, y, t) of the two lines is the collision point. When the detection density is high, the collision point can be clearly obtained, as shown in Figure 2a, but it will lead to excessive computational load because the grid occupancy of the geometric contour of each sampling point on each trajectory needs to be calculated. When the detection density is low, there will be missed detections, as shown in Figure 2b. Although the two trajectories intersect at time t, no sampling is performed at time t due to the sparse sampling points, thus the collision is not detected.

[0068] To address the aforementioned issues, this application provides a collision risk assessment method. This method transforms the vehicle's trajectory to be assessed and the predicted trajectories of other vehicles within a preset range into the same ST coordinate system. Based on the trajectory to be assessed and the predicted trajectories in the ST coordinate system, it determines whether the vehicle faces a collision risk. In this application, the ST coordinate system represents the correspondence between the vehicle's travel distance and time in the current lane, and the vehicle's trajectory in the ST coordinate system appears in a rectangular form, resolving the contradiction between detection density and detection accuracy. A detailed description is provided below with reference to the accompanying drawings and specific embodiments.

[0069] Referring to Figure 3, it is a schematic flowchart of a collision risk assessment method provided in an embodiment of this application. As shown in Figure 3, it mainly includes the following steps.

[0070] Step S301: Obtain the trajectory to be evaluated of the vehicle and the predicted trajectories of other vehicles within the preset range.

[0071] Specifically, the collision risk assessment device can obtain the trajectory of the vehicle to be assessed and the predicted trajectories of other vehicles within a preset range from other modules. In this embodiment, the preset range is the vehicle's perception range. Because the vehicle's perception range is limited, the vehicle cannot obtain the predicted trajectories of vehicles outside its perception range.

[0072] Of course, in practical applications, it can be determined that there is no collision risk for vehicles outside the preset range, so the preset range can be smaller than the vehicle's perception range. This application embodiment does not impose specific restrictions on this.

[0073] Step S302: Convert the trajectory to be evaluated and the predicted trajectory to the same ST coordinate system.

[0074] Specifically, since the trajectory to be evaluated and the predicted trajectory are usually in the XY Cartesian coordinate system, in order to determine the spatiotemporal relationship between the two trajectories more reliably and efficiently, this application first transforms the trajectory to be evaluated and the predicted trajectory in the XY coordinate system to the ST coordinate system to obtain the trajectory diagram in the ST coordinate system, where S is the distance traveled by the vehicle in the current lane and T is the time.

[0075] It is understandable that different driving decisions correspond to different trajectories, and the trajectory to be evaluated in this application embodiment is one of a variety of different trajectories.

[0076] For ease of understanding, this application provides a schematic diagram of the trajectory to be evaluated and the predicted trajectory in the ST coordinate system.

[0077] [Corrected according to Rule 91, 13.01.2026] Refer to Figure 4, which is a schematic diagram of the trajectory to be evaluated and the predicted trajectory in the ST coordinate system provided in an embodiment of this application. As shown in Figure 4, vehicle C1 is another vehicle, vehicle C2 is the own vehicle, the dashed line L1 corresponding to vehicle C1 is the driving trajectory of the other vehicle, the dashed line L2 corresponding to vehicle C2 is the driving trajectory of the own vehicle, the rectangle corresponding to vehicle C1 is the predicted trajectory ST projection, and the rectangle corresponding to vehicle C2 is the trajectory to be evaluated ST projection, and so on in the following text, and will not be repeated. As shown on the left side of Figure 4, lane R1 is the current lane of the own vehicle, and other vehicles are driving in front of the own vehicle. When obtaining the predicted trajectory of other vehicles, the collision risk assessment device can predict the driving distance of other vehicles in the next 6 seconds based on the current driving speed of other vehicles, thereby obtaining the predicted trajectory ST projection in the ST coordinate system on the right side of Figure 4. At the same time, the collision risk assessment device obtains the driving decision of the own vehicle in the next 6 seconds from other modules of the own vehicle, and determines the relationship between the driving distance of the own vehicle in the next 6 seconds and time based on the driving decision, that is, the trajectory ST projection to be evaluated. It is understandable that the slope of the long side of each rectangle in the ST diagram on the right side of Figure 4 is the speed of the corresponding vehicle.

[0078] It should be noted that the 6 seconds, the vehicle speed, the speed of other vehicles, and the number of other vehicles shown in Figure 4 are merely illustrative descriptions and should not be construed as limiting the scope of protection of this application.

[0079] Step S303: Determine whether the vehicle has a collision risk based on the trajectory to be evaluated and the predicted trajectory in the ST coordinate system.

[0080] Specifically, after obtaining the trajectory to be evaluated and the predicted trajectory in the ST coordinate system, the collision risk assessment device determines whether the vehicle has a collision risk based on the ST coordinate system.

[0081] In one possible implementation, it is determined whether the collision detection unit of the trajectory to be evaluated and the collision detection unit of the predicted trajectory overlap in the same time period in the ST coordinate system, where the collision detection unit is the trajectory within a preset time period; if the collision detection unit of the trajectory to be evaluated and the collision detection unit of the predicted trajectory overlap in the same time period in the ST coordinate system, it is determined that the vehicle has a collision risk; if the collision detection unit of the trajectory to be evaluated and the collision detection unit of the predicted trajectory do not overlap in the same time period in the ST coordinate system, it is determined that the vehicle has no collision risk.

[0082] In this embodiment, the collision detection unit is a trajectory within a preset time period. This trajectory can be either the trajectory to be evaluated or the predicted trajectory. It can be understood that the trajectory within the preset time period in the trajectory to be evaluated is the collision detection unit for the trajectory to be evaluated, and the trajectory within the preset time period in the predicted trajectory is the collision detection unit for the predicted trajectory. The preset time period can be any length, and this embodiment does not impose any specific limitations.

[0083] However, because this application applies to multi-step decision-making algorithms such as Partially Observable Markov Decision Processes (POMDPs), and these algorithms require multiple branch decision steps from the initial state to triggering the termination condition and exiting the solution process, in order to obtain a more accurate collision detection unit, the preset time in the embodiments of this application is the minimum step update time.

[0084] Therefore, before determining whether collision detection units overlap, it is necessary to calculate the minimum step update time. Specifically, the minimum step update time is calculated based on the preset decision cycle and the maximum number of decision updates, where the maximum number of decision updates is the maximum number of times the decision can be changed within the preset decision cycle.

[0085] For example, assuming a POMDP behavioral decision-making model with a maximum decision cycle of 6 seconds and a maximum number of decision updates Dmax of 3, then the minimum step update time Δt is 2 seconds. The physical meaning of the maximum number of decision updates Dmax is the number of vertical or lateral behavioral changes within the entire decision cycle. Vertical behaviors include acceleration and deceleration, while lateral behaviors include left turns, right turns, and going straight.

[0086] Assuming the minimum step update time is Δt, and t2 = t1 + Δt, the trajectories to be evaluated corresponding to t1 are St1min and St1max, and the trajectories to be evaluated corresponding to t2 are St2min and St2max. Then, the rectangles with vertices St1min, St1max, St2min, and St2max are the collision detection units for the trajectories to be evaluated. For example, as shown in Figure 4, the minimum step update time is 2 seconds. The collision detection units for the trajectories to be evaluated within 0-2 seconds are the blue rectangles corresponding to 0-2 seconds, and the collision detection units for the predicted trajectories within 0-2 seconds are the red rectangles corresponding to 0-2 seconds.

[0087] In one possible implementation, after obtaining the number of decision updates, the driving style can be determined based on the number of decision updates, where the number of decision updates is the number of times the decision is changed within a preset decision cycle; and the driving decision can be adjusted according to the driving style.

[0088] It should be noted that the maximum number of decision updates corresponds to the maximum decision period, while the number of decision updates corresponds to any preset decision period. In other words, when the preset decision period is the maximum decision period, the number of decision updates is the maximum number of decision updates.

[0089] Understandably, a high number of decision updates indicates frequent changes in driving behavior within the preset decision cycle, such as constant acceleration and deceleration, suggesting an aggressive driving style that can negatively impact passenger comfort. Conversely, a low number of decision updates suggests minimal behavioral changes within the preset decision cycle. In extreme cases, a single decision update might indicate that the driver consistently performs only one of the following longitudinal actions: acceleration, deceleration, or constant speed, or only one of the following lateral actions: left turn or right turn. Combined with the principle of prioritizing driving safety, this results in a conservative and inefficient driving style.

[0090] For example, the trajectory to be evaluated shown in Figure 4 corresponds to a constant speed, meaning the number of decision updates is 1. As can be seen from Figure 4, there is no risk of collision between the vehicle and other vehicles, and the insufficient number of decision updates leads to the vehicle becoming increasingly farther away from other vehicles. Users typically expect vehicles to reach their destination as quickly as possible while ensuring safety, so an insufficient number of decision updates also negatively impacts the user experience.

[0091] This application also provides a schematic diagram of the trajectory to be evaluated and the predicted trajectory in another ST coordinate system.

[0092] Referring to Figure 5, it is a schematic diagram of the trajectory to be evaluated and the predicted trajectory in another ST coordinate system provided in this application embodiment. As shown in Figure 5, the collision detection unit of the trajectory to be evaluated and the collision detection unit of the predicted trajectory do not overlap, so there is no collision risk for the vehicle. Moreover, the vehicle speed is fast at first and then slows down. Compared with Figure 4, the distance traveled in the same amount of time is longer, so the driving efficiency is higher.

[0093] In one possible implementation, the driving style (aggressive or conservative) is determined by comparing the number of decision updates with a preset number, allowing for adjustments to driving decisions based on the determined driving style. In other words, when the driving style is aggressive, the number of decision updates is appropriately reduced; when the driving style is conservative, the number of decision updates is appropriately increased. Of course, increasing the number of decision updates is predicated on ensuring driving safety.

[0094] Although the driving style is conservative when there are fewer decision updates, conservative does not mean that a collision will not occur.

[0095] For example, see Figure 6, which is a schematic diagram of the trajectory to be evaluated and the predicted trajectory in another ST coordinate system provided in an embodiment of this application. As shown in Figure 6, the driving style of this vehicle is conservative, but due to excessive speed, the collision detection unit of the trajectory to be evaluated overlaps with the collision detection unit of the predicted trajectory within 2-4 seconds. That is to say, if the vehicle continues to drive according to the current driving decision, it will collide with other vehicles within 2-4 seconds. This obviously violates the principle of driving safety, so it is necessary to adjust the driving decision in advance to avoid collisions as much as possible.

[0096] Specifically, if the number of decision updates is greater than or equal to a first preset number, the driving style is determined to be aggressive; if the number of decision updates is less than or equal to a second preset number, the driving style is determined to be conservative; wherein the first preset number is greater than the second preset number. Those skilled in the art can set the first and second preset numbers to any number according to actual needs, and this application embodiment does not impose specific limitations on this.

[0097] In one possible implementation, it is determined whether the distance between the predicted trajectory and the trajectory to be evaluated at the same moment is less than or equal to a preset distance threshold. If the distance between the predicted trajectory and the trajectory to be evaluated at the same moment is less than or equal to the preset distance threshold, it is further determined whether the collision detection unit of the trajectory to be evaluated and the collision detection unit of the predicted trajectory in the same time period in the ST coordinate system coincide. If the distance between the predicted trajectory and the trajectory to be evaluated at the same moment is greater than the preset distance threshold, it is determined that there is no collision risk for the vehicle. Those skilled in the art can set the preset distance threshold any number of times according to actual needs, and the embodiments of this application do not impose specific limitations on this.

[0098] [Corrected according to Rule 91, 13.01.2026] For example, taking a side collision as an example, see Figure 7, which is a schematic diagram of the trajectory to be evaluated and the predicted trajectory in another ST coordinate system provided in this application embodiment. As shown in Figure 7, the distance between the predicted trajectory and the trajectory to be evaluated in the 0-2 second interval is greater than a preset distance threshold. At this time, it can be directly determined that the vehicle does not have a collision risk, thereby avoiding some unnecessary computational load. Moreover, if the vehicle drives according to the current driving decision, it will collide with other vehicles within 2-4 seconds. This obviously violates the principle of driving safety, so it is necessary to adjust the driving decision in advance when the predicted trajectory of other vehicles is detected to avoid collisions as much as possible.

[0099] In summary, the risk of collision between a vehicle and other vehicles within a certain time interval Δt can be characterized by a formula, specifically:

[0100] Among them, collision △tUsed to characterize the collision risk within time interval Δt, a collision will occur if the distance l between the predicted trajectory and the trajectory to be evaluated is less than or equal to a preset distance threshold, and there is an intersection between the collision detection units of the trajectory to be evaluated and the collision detection units of the predicted trajectory. Otherwise, no collision will occur. l≤l threshold The distance l between the predicted trajectory and the trajectory to be evaluated is less than or equal to a preset distance threshold l. threshold , Collision detection unit characterizing the trajectory to be evaluated Collision detection unit with predicted trajectory There is an intersection.

[0101] In summary, the embodiments of this application utilize the characteristics of multi-step decision-making algorithms such as POMDP to construct a more flexible collision detection unit according to the decision step size, thereby making it easier to plan a safe and highly efficient driving trajectory.

[0102] Corresponding to the above embodiments, this application also provides a collision risk assessment device.

[0103] Referring to Figure 8, a schematic diagram of a collision risk assessment device provided in an embodiment of this application is shown. As shown in Figure 8, the collision risk assessment device includes a trajectory acquisition module 801, a coordinate transformation module 802, and a collision risk determination module 803. These components communicate through one or more buses. Those skilled in the art will understand that the structure of the electronic device shown in the figure does not constitute a limitation on the embodiment of this application. It can be a bus-shaped structure or a star-shaped structure, and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0104] Among them, the trajectory acquisition module 801 is used to acquire the trajectory to be evaluated of the vehicle and the predicted trajectories of other vehicles within a preset range;

[0105] The coordinate transformation module 802 is used to transform the trajectory to be evaluated and the predicted trajectory to the same ST coordinate system, where S is the distance traveled by the vehicle in the current lane and T is the time.

[0106] The collision risk determination module 803 is used to determine whether the vehicle has a collision risk based on the trajectory to be evaluated and the predicted trajectory in the ST coordinate system.

[0107] Corresponding to the above embodiments, this application also provides an electronic device.

[0108] Referring to Figure 9, a schematic diagram of the structure of an electronic device provided in an embodiment of this application is shown. As shown in Figure 9, the electronic device 900 may include a processor 901, a memory 902, and a communication unit 903. These components communicate through one or more buses. Those skilled in the art will understand that the structure of the electronic device shown in the figure does not constitute a limitation on the embodiment of this application. It may be a bus topology or a star topology, and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0109] The communication unit 903 is used to establish a communication channel, enabling the electronic device to communicate with other devices. It can receive user data sent by other devices or send user data to other devices.

[0110] The processor 901 serves as the control center of the electronic device, connecting various parts of the device via interfaces and lines. It executes software programs, instructions, and / or modules stored in the memory 902, and calls data stored in the memory to perform various functions and / or process data. The processor may be composed of integrated circuits (ICs), such as a single packaged IC or multiple packaged ICs with the same or different functions connected together. For example, the processor 901 may consist only of a central processing unit (CPU). In this embodiment, the CPU may have a single processing core or include multiple processing cores.

[0111] The memory 902 is used to store the execution instructions of the processor 901. The memory 902 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk or optical disk.

[0112] When the execution instructions in memory 902 are executed by processor 901, the electronic device 900 is able to perform some or all of the steps in the embodiment shown in FIG3.

[0113] In a specific implementation, this application embodiment also provides a computer storage medium, wherein the computer storage medium may store a program, and when the program is executed, it may include some or all of the steps of the simulation scene generation method provided in various embodiments of this application. The storage medium may be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.

[0114] In this application embodiment, "at least one" refers to one or more, and "more than one" refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent the existence of A alone, the simultaneous existence of A and B, or the existence of B alone. A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one of the following" and similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one of a, b, and c can represent: a, b, c, ab, ac, bc, or abc, where a, b, and c can be single or multiple.

[0115] Those skilled in the art will recognize that the units and algorithm steps described in the embodiments disclosed herein can be implemented using electronic hardware, computer software, or a combination of electronic hardware and software. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0116] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0117] In the several embodiments provided in this application, any function, if implemented as a software functional unit and sold or used as an independent product, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0118] The same or similar parts between the various embodiments in this specification can be referred to mutually. In particular, the device embodiments and terminal embodiments are basically similar to the method embodiments, so the description is relatively simple, and the relevant parts can be referred to the description in the method embodiments.

Claims

1. A collision risk assessment method, characterized in that, include: Acquire the trajectory of the vehicle to be evaluated and the predicted trajectories of other vehicles within a preset range; The trajectory to be evaluated and the predicted trajectory are transformed to the same ST coordinate system, where S is the distance traveled by the vehicle in the current lane and T is the time. Based on the trajectory to be evaluated and the predicted trajectory in the ST coordinate system, it is determined whether the vehicle has a collision risk.

2. The method according to claim 1, characterized in that, The step of determining whether the vehicle has a collision risk based on the trajectory to be evaluated and the predicted trajectory in the ST coordinate system includes: Determine whether the collision detection unit of the trajectory to be evaluated and the collision detection unit of the predicted trajectory coincide in the same time period in the ST coordinate system, wherein the collision detection unit is the trajectory within a preset time period; If the collision detection unit of the trajectory to be evaluated and the collision detection unit of the predicted trajectory overlap in the same time period in the ST coordinate system, then it is determined that the vehicle has a collision risk. If the collision detection unit of the trajectory to be evaluated and the collision detection unit of the predicted trajectory do not overlap in the same time period in the ST coordinate system, then it is determined that the vehicle has no collision risk.

3. The method according to claim 2, characterized in that, The preset duration is the minimum step update duration. Before determining whether the collision detection unit of the trajectory to be evaluated and the collision detection unit of the predicted trajectory overlap in the same time period in the ST coordinate system, the method further includes: The minimum step update time is calculated based on the preset decision cycle and the maximum number of decision updates, where the maximum number of decision updates is the maximum number of times the decision is changed within the preset decision cycle.

4. The method according to claim 3, characterized in that, The method further includes: The driving style is determined based on the number of decision updates, where the number of decision updates is the number of times the decision is changed within the preset decision period; Adjust driving decisions based on the described driving style.

5. The method according to claim 4, characterized in that, The step of determining driving style based on the number of decision updates includes: If the number of decision updates is greater than or equal to the first preset number, then the driving style is determined to be aggressive. If the number of decision updates is less than or equal to the second preset number, then the driving style is determined to be conservative. The first preset number of times is greater than the second preset number of times.

6. The method according to claim 2, characterized in that, The step of determining whether the collision detection unit of the trajectory to be evaluated and the collision detection unit of the predicted trajectory coincide in the same time period in the ST coordinate system includes: If the distance between the predicted trajectory and the trajectory to be evaluated at the same moment is less than or equal to a preset distance threshold, then it is determined whether the collision detection unit of the trajectory to be evaluated and the collision detection unit of the predicted trajectory in the same time period in the ST coordinate system overlap.

7. The method according to claim 6, characterized in that, The method further includes: Determine whether the distance between the predicted trajectory and the trajectory to be evaluated at the same moment is less than or equal to a preset distance threshold; If the distance between the predicted trajectory and the trajectory to be evaluated is greater than a preset distance threshold at the same time, it is determined that the vehicle does not have a collision risk.

8. A collision risk assessment device, characterized in that, include: The trajectory acquisition module is used to acquire the trajectory to be evaluated of the vehicle and the predicted trajectories of other vehicles within a preset range; The coordinate transformation module is used to transform the trajectory to be evaluated and the predicted trajectory to the same ST coordinate system, where S is the distance traveled by the vehicle in the current lane and T is the time. The collision risk determination module is used to determine whether the vehicle has a collision risk based on the trajectory to be evaluated and the predicted trajectory in the ST coordinate system.

9. An electronic device, characterized in that, include: processor; Memory; And a computer program, wherein the computer program is stored in the memory, the computer program including instructions that, when executed by the processor, cause the electronic device to perform the method of any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored program, wherein, when the program is executed, it controls the device on which the computer-readable storage medium is located to perform the method according to any one of claims 1 to 7.

Images