Vehicle collision probability detection method and device

Abstract

The application relates to a vehicle collision probability detection method and device, which is applied to the field of automatic driving, intelligent driving or unmanned driving. The method comprises the following steps: acquiring state information and a reference line of a current vehicle, and acquiring state information and a reference line of at least one target vehicle; predicting a driving track of the current vehicle according to the state information of the current vehicle; predicting a driving track of the at least one target vehicle according to the state information of the at least one target vehicle; acquiring a planned track of the current vehicle and a planned track of the at least one target vehicle according to the state information and the reference line of the current vehicle and the state information and the reference line of the at least one target vehicle; and the fusion of the predicted driving track and the planned track indicates the collision probability between the current vehicle and each target vehicle and the collision probability between each target vehicle. The embodiment of the present application can improve the accuracy of vehicle collision probability detection.

Description

Technical Field

[0001] This application relates to the field of intelligent transportation technology, and in particular to a method and apparatus for vehicle collision probability detection. Background Technology

[0002] When a driver is driving on the road, in the event of an impending collision with surrounding vehicles, the driver's delayed reaction time can easily lead to a collision or a chain reaction of collisions, resulting in a serious traffic accident. Existing collision risk detection or collision probability calculations typically only consider the behavior of the driver and surrounding vehicles, resulting in a relatively inaccurate collision probability (CP) that cannot effectively reduce the incidence of collision accidents. Summary of the Invention

[0003] In view of this, a method, device and storage medium for vehicle collision probability detection are proposed.

[0004] In a first aspect, embodiments of this application provide a vehicle collision probability detection method, the method comprising: acquiring the state information and reference line of a current vehicle, and the state information and reference line of at least one target vehicle, wherein the state information includes at least one of lateral position, longitudinal position, velocity, acceleration, heading angle, and heading angular velocity, the reference line of the current vehicle includes the center line of the lane in which the current vehicle is located, and the at least one target vehicle is a vehicle surrounding the current vehicle; predicting the driving trajectory of the current vehicle based on the state information of the current vehicle; predicting the driving trajectory of the at least one target vehicle based on the state information and reference line of the current vehicle, and the state information and reference line of the at least one target vehicle; and acquiring the planned trajectory of the current vehicle and the planned trajectory of the at least one target vehicle based on the state information and reference line of the current vehicle and the at least one target vehicle, wherein the fusion of the predicted driving trajectory and the planned trajectory indicates the collision probability between the current vehicle and each target vehicle and the collision probability between each target vehicle.

[0005] According to embodiments of this application, it is possible to obtain the status information and reference lines of the current vehicle (self-vehicle), and the status information and reference lines of at least one target vehicle (vehicles surrounding the self-vehicle). Based on the status information of the current vehicle, the driving trajectory of the current vehicle is predicted. Based on the status information of at least one target vehicle, the driving trajectory of at least one target vehicle is predicted. Based on the status information and reference lines of the current vehicle and at least one target vehicle, the planned trajectory of the current vehicle and the planned trajectory of at least one target vehicle are obtained. The fusion of the predicted driving trajectory and the planned trajectory indicates the collision probability between the current vehicle and each target vehicle, as well as the collision probability between each target vehicle. This can improve the accuracy of vehicle collision probability detection, thereby effectively reducing the incidence of vehicle collision accidents and improving the driving safety of the self-vehicle.

[0006] According to the first aspect, in a first possible implementation of the vehicle collision probability detection method, the method further includes: fusing the predicted driving trajectory of the current vehicle with the planned trajectory to obtain a fused trajectory of the current vehicle; fusing the predicted driving trajectory of the at least one target vehicle with the planned trajectory to obtain a fused trajectory of the at least one target vehicle; and determining the collision probability between the current vehicle and each target vehicle, and the collision probability between each target vehicle, based on the fused trajectory of the current vehicle and the fused trajectory of the at least one target vehicle.

[0007] According to embodiments of this application, a fused trajectory of the current vehicle is obtained by fusing the predicted driving trajectory of the current vehicle with the planned trajectory, and a fused trajectory of at least one target vehicle is obtained by fusing the predicted driving trajectory of at least one target vehicle with the planned trajectory. Based on the fused trajectory of the current vehicle and the fused trajectory of at least one target vehicle, the collision probability between the current vehicle and each target vehicle and the collision probability between each target vehicle are determined. This allows the fusion of the predicted driving trajectory with the planned trajectory that takes into account obstacle avoidance, resulting in a fused trajectory that is more in line with the behavior habits of rational drivers. The collision probability between the vehicle and surrounding vehicles and the collision probability between each surrounding vehicle are determined based on this fused trajectory. This makes the collision probability calculation consider not only the behavior of the vehicle and surrounding vehicles, but also the behavior of surrounding vehicles with each other, thereby improving the accuracy of vehicle collision probability detection, effectively reducing the incidence of vehicle collision accidents, and improving the driving safety of the vehicle.

[0008] According to the first aspect or the first possible implementation of the first aspect, in the second possible implementation of the vehicle collision probability detection method, the method further includes: outputting prompt information, the prompt information including at least one of voice prompt information and visual prompt information, the prompt information being used to indicate the collision probability.

[0009] According to embodiments of this application, voice, visual, and other prompts can be output within the cabin of the current vehicle based on the collision probability, thereby increasing the driver's alertness and improving the vehicle's driving safety.

[0010] According to the first aspect, or the first possible implementation of the first aspect, or the second possible implementation of the first aspect, in a third possible implementation of the vehicle collision probability detection method, the method further includes: performing corresponding driving control on the current vehicle according to the collision probability.

[0011] According to embodiments of this application, by performing corresponding driving control on the current vehicle based on the collision probability, it is possible not only to reduce the incidence of vehicle collision accidents, but also to improve the driving safety of the vehicle.

[0012] According to any one of the first aspect or the first to third possible implementations of the first aspect, in a fourth possible implementation of the vehicle collision probability detection method, the step of determining the collision probability between the current vehicle and each target vehicle, and between each target vehicle, based on the fused trajectory of the current vehicle and the fused trajectory of the at least one target vehicle, includes: determining a trajectory set of a first vehicle, wherein the first vehicle is any one of the current vehicle and the at least one target vehicle, the trajectory set includes at least one trajectory combination, each trajectory combination including the fused trajectory of the first vehicle and the fused trajectory of a second vehicle, wherein the second vehicle is any one of the current vehicle and the at least one target vehicle other than the first vehicle; determining the cost value of each trajectory combination in the trajectory set; and determining the collision probability between the first vehicle and the second vehicle based on the cost value of each trajectory combination in the trajectory set.

[0013] According to an embodiment of this application, by establishing a trajectory set of a first vehicle and determining the cost value of each trajectory combination in the trajectory set, the collision probability between the first vehicle and the second vehicle can be determined based on the cost value of each trajectory combination in the trajectory set, thereby improving the accuracy of the collision probability.

[0014] According to the fourth possible implementation of the first aspect, in the fifth possible implementation of the vehicle collision probability detection method, determining the cost value of each trajectory combination in the trajectory set includes: determining whether each trajectory combination in the trajectory set has collided; determining the trajectory combinations that have not collided as a non-collision trajectory set; determining the cost value of each trajectory combination in the non-collision trajectory set; and determining the collision probability between the first vehicle and the second vehicle based on the cost value of each trajectory combination in the trajectory set includes: determining the collision probability between the first vehicle and the second vehicle based on the cost value of each trajectory combination in the non-collision trajectory set.

[0015] According to the embodiments of this application, by collision judgment, non-collision trajectory combinations are selected from the trajectory set of the first vehicle to establish a non-collision trajectory set, and the cost value of each trajectory combination in the non-collision trajectory set is determined. Then, based on the cost value of each trajectory combination in the non-collision trajectory set, the collision probability between the first vehicle and the second vehicle is determined. This can reduce the number of trajectory combinations for calculating cost values, thereby improving processing efficiency.

[0016] According to the fifth possible implementation of the first aspect, in the sixth possible implementation of the vehicle collision probability detection method, determining whether a collision occurs in each trajectory combination in the trajectory set includes: for any trajectory combination, discretizing the fused trajectory of the first vehicle and the fused trajectory of the second vehicle in the trajectory combination according to a preset first time series to obtain multiple target positions of the first vehicle and multiple target positions of the second vehicle, wherein the first time series includes multiple moments; at any moment in the first time series, determining whether there is an overlap between the target area where the first vehicle is located and the target area where the second vehicle is located, wherein the target area where the first vehicle is located corresponds to the target position of the first vehicle at that moment, and the target area where the second vehicle is located corresponds to the target position of the second vehicle at that moment, wherein the overlap of the target areas indicates that a collision has occurred in the trajectory combination.

[0017] According to embodiments of this application, determining whether a trajectory combination collides by overlapping regions can improve the accuracy of collision determination for trajectory combinations.

[0018] According to any one of the first aspect or the first to the sixth possible implementations of the first aspect, in the seventh possible implementation of the vehicle collision probability detection method, the target area is a rectangular area or a three-circle area, wherein the three-circle area includes three circular areas defined by the center of the front axle, the center of the vehicle, and the center of the rear axle of the vehicle as the center and the wheelbase of the vehicle as the diameter.

[0019] According to embodiments of this application, setting the target area as a rectangular area or a three-circle area can improve the conformity between the target area and the vehicle's shape, thereby improving the accuracy of the vehicle's target area and thus improving the accuracy of collision judgment based on trajectory combinations.

[0020] According to the first possible implementation of the first aspect, in the eighth possible implementation of the vehicle collision probability detection method, the step of fusing the predicted driving trajectory and the planned trajectory of the current vehicle to obtain the fused trajectory of the current vehicle includes: discretizing the predicted driving trajectory and the planned trajectory of the current vehicle according to a preset second time series to obtain multiple predicted trajectory points and multiple planned trajectory points; and determining the fused trajectory of the current vehicle according to a preset weight sequence, the multiple predicted trajectory points and the multiple planned trajectory points.

[0021] According to an embodiment of this application, the predicted driving trajectory and the planned trajectory of the current vehicle are fused by point fusion to obtain the fused trajectory of the current vehicle, which can improve the accuracy of the fused trajectory.

[0022] According to the first possible implementation of the first aspect, in the ninth possible implementation of the vehicle collision probability detection method, fusing the predicted driving trajectory and the planned trajectory of the current vehicle to obtain the fused trajectory of the current vehicle includes: determining the first polynomial coefficient corresponding to the predicted driving trajectory of the current vehicle and the second polynomial coefficient corresponding to the planned trajectory of the current vehicle; and determining the fused trajectory of the current vehicle according to the preset coefficient weights, the first polynomial coefficients and the second polynomial coefficients.

[0023] According to the embodiments of this application, the predicted driving trajectory and the planned trajectory of the current vehicle are fused by coefficient fusion to obtain the fused trajectory of the current vehicle. This not only improves the accuracy of the fused trajectory, but also reduces the amount of computation and improves processing efficiency.

[0024] According to any one of the first aspect or the first to the ninth possible implementations of the first aspect, in the tenth possible implementation of the vehicle collision probability detection method, the step of predicting the driving trajectory of the current vehicle based on the current vehicle's state information includes: determining at least one future position of the current vehicle based on the current vehicle's state information; and predicting the driving trajectory of the current vehicle based on the at least one future position.

[0025] According to embodiments of this application, the current vehicle's trajectory can be predicted using at least one future location of the current vehicle, which is simple and fast, thereby improving processing efficiency.

[0026] According to any one of the first aspect or the first to the ninth possible implementations of the first aspect, in the eleventh possible implementation of the vehicle collision probability detection method, the step of predicting the driving trajectory of the current vehicle based on the current vehicle's state information includes: obtaining at least one historical position of the current vehicle; determining at least one future position of the current vehicle based on the current vehicle's state information; and predicting the driving trajectory of the current vehicle based on the at least one historical position and the at least one future position.

[0027] According to embodiments of this application, the driving trajectory of the current vehicle is predicted by using at least one historical location and at least one future location of the current vehicle, thereby improving the accuracy of the predicted driving trajectory.

[0028] According to any one of the first aspect or the first to eleventh possible implementations of the first aspect, in the twelfth possible implementation of the vehicle collision probability detection method, the step of obtaining the planned trajectory of the current vehicle and the planned trajectory of the at least one target vehicle based on the state information and reference line of the current vehicle and the state information and reference line of the at least one target vehicle includes: for the current vehicle, determining at least one planned position of the current vehicle based on the state information and reference line of the current vehicle, the state information and reference line of the at least one target vehicle, and a preset position interval and speed interval; and determining the planned trajectory of the current vehicle based on the at least one planned position.

[0029] According to embodiments of this application, at least one planned position of the current vehicle is determined by the current vehicle's status information and reference line, the status information and reference line of at least one target vehicle, and preset position and speed intervals. Based on the at least one planned position, the planned trajectory of the current vehicle is determined, thereby improving the accuracy of the planned trajectory of the current vehicle.

[0030] Secondly, embodiments of this application provide a vehicle collision probability detection device. The device includes: an acquisition unit, configured to acquire state information and a reference line of a current vehicle, and state information and a reference line of at least one target vehicle. The state information includes at least one of lateral position, longitudinal position, velocity, acceleration, heading angle, and heading angular velocity. The reference line of the current vehicle includes the center line of the lane in which the current vehicle is located. The at least one target vehicle is a vehicle surrounding the current vehicle. A processing unit is configured to: predict the driving trajectory of the current vehicle based on the state information of the current vehicle; predict the driving trajectory of the at least one target vehicle based on the state information of the at least one target vehicle; and acquire the planned trajectory of the current vehicle and the planned trajectory of the at least one target vehicle based on the state information and reference line of the current vehicle and the state information and reference line of the at least one target vehicle. The fusion of the predicted driving trajectory and the planned trajectory indicates the collision probability between the current vehicle and each target vehicle, and the collision probability between each target vehicle.

[0031] According to embodiments of this application, it is possible to obtain the status information and reference lines of the current vehicle (self-vehicle), and the status information and reference lines of at least one target vehicle (vehicles surrounding the self-vehicle). Based on the status information of the current vehicle, the driving trajectory of the current vehicle is predicted. Based on the status information of at least one target vehicle, the driving trajectory of at least one target vehicle is predicted. Based on the status information and reference lines of the current vehicle and at least one target vehicle, the planned trajectory of the current vehicle and the planned trajectory of at least one target vehicle are obtained. The fusion of the predicted driving trajectory and the planned trajectory indicates the collision probability between the current vehicle and each target vehicle, as well as the collision probability between each target vehicle. This can improve the accuracy of vehicle collision probability detection, thereby effectively reducing the incidence of vehicle collision accidents and improving the driving safety of the self-vehicle.

[0032] According to the second aspect, in a first possible implementation of the vehicle collision probability detection device, the processing unit is further configured to: fuse the predicted driving trajectory of the current vehicle with the planned trajectory to obtain a fused trajectory of the current vehicle; fuse the predicted driving trajectory of the at least one target vehicle with the planned trajectory to obtain a fused trajectory of the at least one target vehicle; and determine the collision probability between the current vehicle and each target vehicle, and the collision probability between each target vehicle, based on the fused trajectory of the current vehicle and the fused trajectory of the at least one target vehicle.

[0033] According to embodiments of this application, a fused trajectory of the current vehicle is obtained by fusing the predicted driving trajectory of the current vehicle with the planned trajectory, and a fused trajectory of at least one target vehicle is obtained by fusing the predicted driving trajectory of at least one target vehicle with the planned trajectory. Based on the fused trajectory of the current vehicle and the fused trajectory of at least one target vehicle, the collision probability between the current vehicle and each target vehicle and the collision probability between each target vehicle are determined. This allows the fusion of the predicted driving trajectory with the planned trajectory that takes into account obstacle avoidance, resulting in a fused trajectory that is more in line with the behavior habits of rational drivers. The collision probability between the vehicle and surrounding vehicles and the collision probability between each surrounding vehicle are determined based on this fused trajectory. This makes the collision probability calculation consider not only the behavior of the vehicle and surrounding vehicles, but also the behavior of surrounding vehicles with each other, thereby improving the accuracy of vehicle collision probability detection, effectively reducing the incidence of vehicle collision accidents, and improving the driving safety of the vehicle.

[0034] According to the second aspect or the first possible implementation of the second aspect, in the second possible implementation of the vehicle collision probability detection device, the device further includes: an output unit for outputting prompt information, the prompt information including at least one of voice prompt information and visual prompt information, the prompt information being used to indicate the collision probability.

[0035] According to embodiments of this application, voice, visual, and other prompts can be output within the cabin of the current vehicle based on the collision probability, thereby increasing the driver's alertness and improving the vehicle's driving safety.

[0036] According to the second aspect or the first or second possible implementation of the second aspect, in a third possible implementation of the vehicle collision probability detection device, the device further includes: a control unit, which performs corresponding driving control on the current vehicle according to the collision probability.

[0037] According to embodiments of this application, by performing corresponding driving control on the current vehicle based on the collision probability, it is possible not only to reduce the incidence of vehicle collision accidents, but also to improve the driving safety of the vehicle.

[0038] According to any one of the second aspect or the first to third possible implementations of the second aspect, in a fourth possible implementation of the vehicle collision probability detection device, the step of determining the collision probability between the current vehicle and each target vehicle, and between each target vehicle, based on the fused trajectory of the current vehicle and the fused trajectory of the at least one target vehicle, includes: determining a trajectory set of a first vehicle, wherein the first vehicle is any one of the current vehicle and the at least one target vehicle, the trajectory set includes at least one trajectory combination, each trajectory combination including the fused trajectory of the first vehicle and the fused trajectory of a second vehicle, wherein the second vehicle is any one of the current vehicle and the at least one target vehicle other than the first vehicle; determining the cost value of each trajectory combination in the trajectory set; and determining the collision probability between the first vehicle and the second vehicle based on the cost value of each trajectory combination in the trajectory set.

[0039] According to an embodiment of this application, by establishing a trajectory set of a first vehicle and determining the cost value of each trajectory combination in the trajectory set, the collision probability between the first vehicle and the second vehicle can be determined based on the cost value of each trajectory combination in the trajectory set, thereby improving the accuracy of the collision probability.

[0040] According to the fourth possible implementation of the second aspect, in the fifth possible implementation of the vehicle collision probability detection device, determining the cost value of each trajectory combination in the trajectory set includes: determining whether each trajectory combination in the trajectory set has collided; determining the trajectory combinations that have not collided as a non-collision trajectory set; determining the cost value of each trajectory combination in the non-collision trajectory set; and determining the collision probability between the first vehicle and the second vehicle based on the cost value of each trajectory combination in the trajectory set includes: determining the collision probability between the first vehicle and the second vehicle based on the cost value of each trajectory combination in the non-collision trajectory set.

[0041] According to the embodiments of this application, by collision judgment, non-collision trajectory combinations are selected from the trajectory set of the first vehicle to establish a non-collision trajectory set, and the cost value of each trajectory combination in the non-collision trajectory set is determined. Then, based on the cost value of each trajectory combination in the non-collision trajectory set, the collision probability between the first vehicle and the second vehicle is determined. This can reduce the number of trajectory combinations for calculating cost values, thereby improving processing efficiency.

[0042] According to the fifth possible implementation of the second aspect, in the sixth possible implementation of the vehicle collision probability detection device, determining whether a collision occurs in each trajectory combination in the trajectory set includes: for any trajectory combination, discretizing the fused trajectory of the first vehicle and the fused trajectory of the second vehicle in the trajectory combination according to a preset first time sequence to obtain multiple target positions of the first vehicle and multiple target positions of the second vehicle, wherein the first time sequence includes multiple moments; at any moment in the first time sequence, determining whether there is an overlap between the target area where the first vehicle is located and the target area where the second vehicle is located, wherein the target area where the first vehicle is located corresponds to the target position of the first vehicle at that moment, and the target area where the second vehicle is located corresponds to the target position of the second vehicle at that moment, wherein the overlap of the target areas indicates that a collision has occurred in the trajectory combination.

[0043] According to embodiments of this application, determining whether a trajectory combination collides by overlapping regions can improve the accuracy of collision determination for trajectory combinations.

[0044] According to any one of the second aspect or the first to the sixth possible implementations of the second aspect, in the seventh possible implementation of the vehicle collision probability detection device, the target area is a rectangular area or a three-circle area, wherein the three-circle area includes three circular areas defined by the center of the front axle, the center of the vehicle, and the center of the rear axle of the vehicle as the center and the wheelbase of the vehicle as the diameter.

[0045] According to embodiments of this application, setting the target area as a rectangular area or a three-circle area can improve the conformity between the target area and the vehicle's shape, thereby improving the accuracy of the vehicle's target area and thus improving the accuracy of collision judgment based on trajectory combinations.

[0046] According to the first possible implementation of the second aspect, in the eighth possible implementation of the vehicle collision probability detection device, fusing the predicted driving trajectory and the planned trajectory of the current vehicle to obtain the fused trajectory of the current vehicle includes: discretizing the predicted driving trajectory and the planned trajectory of the current vehicle according to a preset second time series to obtain multiple predicted trajectory points and multiple planned trajectory points; and determining the fused trajectory of the current vehicle according to a preset weight sequence, the multiple predicted trajectory points and the multiple planned trajectory points.

[0047] According to an embodiment of this application, the predicted driving trajectory and the planned trajectory of the current vehicle are fused by point fusion to obtain the fused trajectory of the current vehicle, which can improve the accuracy of the fused trajectory.

[0048] According to the first possible implementation of the second aspect, in the ninth possible implementation of the vehicle collision probability detection device, fusing the predicted driving trajectory and the planned trajectory of the current vehicle to obtain the fused trajectory of the current vehicle includes: determining the first polynomial coefficient corresponding to the predicted driving trajectory of the current vehicle and the second polynomial coefficient corresponding to the planned trajectory of the current vehicle; and determining the fused trajectory of the current vehicle according to the preset coefficient weights, the first polynomial coefficients and the second polynomial coefficients.

[0049] According to the embodiments of this application, the predicted driving trajectory and the planned trajectory of the current vehicle are fused by coefficient fusion to obtain the fused trajectory of the current vehicle. This not only improves the accuracy of the fused trajectory, but also reduces the amount of computation and improves processing efficiency.

[0050] According to any one of the second aspect or the first to the ninth possible implementations of the second aspect, in the tenth possible implementation of the vehicle collision probability detection device, predicting the driving trajectory of the current vehicle based on the current vehicle's state information includes: determining at least one future position of the current vehicle based on the current vehicle's state information; and predicting the driving trajectory of the current vehicle based on the at least one future position.

[0051] According to embodiments of this application, the current vehicle's trajectory can be predicted using at least one future location of the current vehicle, which is simple and fast, thereby improving processing efficiency.

[0052] According to any one of the second aspect or the first to the ninth possible implementations of the second aspect, in the eleventh possible implementation of the vehicle collision probability detection device, the step of predicting the driving trajectory of the current vehicle based on the current vehicle's state information includes: acquiring at least one historical position of the current vehicle; determining at least one future position of the current vehicle based on the current vehicle's state information; and predicting the driving trajectory of the current vehicle based on the at least one historical position and the at least one future position.

[0053] According to embodiments of this application, the driving trajectory of the current vehicle is predicted by using at least one historical location and at least one future location of the current vehicle, thereby improving the accuracy of the predicted driving trajectory.

[0054] According to any one of the second aspect or the first to eleventh possible implementations of the second aspect, in the twelfth possible implementation of the vehicle collision probability detection device, the step of obtaining the planned trajectory of the current vehicle and the planned trajectory of the at least one target vehicle based on the current vehicle's state information and reference line, and the state information and reference line of the at least one target vehicle includes: for the current vehicle, determining at least one planned position of the current vehicle based on the current vehicle's state information and reference line, the at least one target vehicle's state information and reference line, and preset position and speed intervals; and determining the planned trajectory of the current vehicle based on the at least one planned position.

[0055] According to an embodiment of this application, at least one planned position of the current vehicle is determined by the current vehicle's status information and reference line, the status information and reference line of at least one target vehicle, and preset position and speed intervals. Based on the at least one planned position, the planned trajectory of the current vehicle is determined, thereby improving the accuracy of the planned trajectory of the current vehicle.

[0056] Thirdly, a vehicle collision probability detection device, the device comprising: a processor; a memory for storing processor-executable instructions; wherein the processor is configured to execute one or more of the vehicle collision probability detection methods described in the first aspect or in various possible implementations of the first aspect.

[0057] According to embodiments of this application, it is possible to obtain the status information and reference lines of the current vehicle (self-vehicle), and the status information and reference lines of at least one target vehicle (vehicles surrounding the self-vehicle). Based on the status information of the current vehicle, the driving trajectory of the current vehicle is predicted. Based on the status information of at least one target vehicle, the driving trajectory of at least one target vehicle is predicted. Based on the status information and reference lines of the current vehicle and at least one target vehicle, the planned trajectory of the current vehicle and the planned trajectory of at least one target vehicle are obtained. The fusion of the predicted driving trajectory and the planned trajectory indicates the collision probability between the current vehicle and each target vehicle, as well as the collision probability between each target vehicle. This can improve the accuracy of vehicle collision probability detection, thereby effectively reducing the incidence of vehicle collision accidents and improving the driving safety of the self-vehicle.

[0058] Fourthly, embodiments of this application provide a non-volatile computer-readable storage medium, the computer-readable storage medium including computer instructions, which, when executed on a computer, cause the computer to perform one or more of the vehicle collision probability detection methods described in the first aspect or various possible implementations of the first aspect.

[0059] According to embodiments of this application, it is possible to obtain the status information and reference lines of the current vehicle (self-vehicle), and the status information and reference lines of at least one target vehicle (vehicles surrounding the self-vehicle). Based on the status information of the current vehicle, the driving trajectory of the current vehicle is predicted. Based on the status information of at least one target vehicle, the driving trajectory of at least one target vehicle is predicted. Based on the status information and reference lines of the current vehicle and at least one target vehicle, the planned trajectory of the current vehicle and the planned trajectory of at least one target vehicle are obtained. The fusion of the predicted driving trajectory and the planned trajectory indicates the collision probability between the current vehicle and each target vehicle, as well as the collision probability between each target vehicle. This can improve the accuracy of vehicle collision probability detection, thereby effectively reducing the incidence of vehicle collision accidents and improving the driving safety of the self-vehicle.

[0060] Fifthly, embodiments of this application provide a computer program product including computer-readable code, or a non-volatile computer-readable storage medium carrying computer-readable code. When the computer-readable code is run in an electronic device, the processor in the electronic device executes one or more of the vehicle collision probability detection methods described in the first aspect or various possible implementations of the first aspect.

[0061] According to embodiments of this application, it is possible to obtain the status information and reference lines of the current vehicle (self-vehicle), and the status information and reference lines of at least one target vehicle (vehicles surrounding the self-vehicle). Based on the status information of the current vehicle, the driving trajectory of the current vehicle is predicted. Based on the status information of at least one target vehicle, the driving trajectory of at least one target vehicle is predicted. Based on the status information and reference lines of the current vehicle and at least one target vehicle, the planned trajectory of the current vehicle and the planned trajectory of at least one target vehicle are obtained. The fusion of the predicted driving trajectory and the planned trajectory indicates the collision probability between the current vehicle and each target vehicle, as well as the collision probability between each target vehicle. This can improve the accuracy of vehicle collision probability detection, thereby effectively reducing the incidence of vehicle collision accidents and improving the driving safety of the self-vehicle.

[0062] Sixthly, embodiments of this application provide a vehicle, the vehicle including a vehicle collision probability detection device, the vehicle collision probability detection device being one or more of the vehicle collision probability detection devices described in the second aspect or multiple possible implementations of the second aspect, or the vehicle collision probability detection device described in the third aspect.

[0063] These and other aspects of this application will become more apparent in the description of the following embodiments(s). Attached Figure Description

[0064] The accompanying drawings, which are included in and form part of this specification, illustrate exemplary embodiments, features, and aspects of this application together with the specification and serve to explain the principles of this application.

[0065] Figure 1 This diagram illustrates a prior art method for calculating collision probability.

[0066] Figure 2 This diagram illustrates a situation where the predicted trajectory of a vehicle deviates significantly from its actual future trajectory.

[0067] Figure 3a , Figure 3b and Figure 3c This diagram illustrates an application scenario of a vehicle collision probability detection method according to an embodiment of this application.

[0068] Figure 4 A flowchart illustrating a vehicle collision probability detection method according to an embodiment of this application is shown.

[0069] Figure 5 A schematic diagram showing a vehicle reference line in a vehicle collision probability detection method according to an embodiment of this application is provided.

[0070] Figure 6 A schematic diagram illustrating coordinate transformation in a vehicle collision probability detection method according to an embodiment of this application is shown.

[0071] Figure 7 A schematic diagram showing the predicted driving trajectory of a vehicle collision probability detection method according to an embodiment of this application is illustrated.

[0072] Figure 8 A schematic diagram showing coordinate rotation and translation of a vehicle collision probability detection method according to an embodiment of this application is provided.

[0073] Figure 9 A schematic diagram showing the planning path of a vehicle collision probability detection method according to an embodiment of this application is provided.

[0074] Figure 10 A flowchart illustrating a vehicle collision probability detection method according to an embodiment of this application is shown.

[0075] Figure 11 A schematic diagram illustrating trajectory fusion in a vehicle collision probability detection method according to an embodiment of this application is shown.

[0076] Figure 12 A schematic diagram showing the target area of ​​a vehicle collision probability detection method according to an embodiment of this application is provided.

[0077] Figure 13 A schematic diagram showing the target area of ​​a vehicle collision probability detection method according to an embodiment of this application is provided.

[0078] Figure 14 A block diagram of a vehicle collision probability detection device according to an embodiment of this application is shown. Detailed Implementation

[0079] Various exemplary embodiments, features, and aspects of this application will now be described in detail with reference to the accompanying drawings. The same reference numerals in the drawings denote elements that have the same or similar functions. Although various aspects of the embodiments are shown in the drawings, they are not necessarily drawn to scale unless specifically indicated otherwise.

[0080] The term “exemplary” as used herein means “serving as an example, embodiment, or illustration.” Any embodiment illustrated herein as “exemplary” is not necessarily to be construed as superior to or better than other embodiments.

[0081] Furthermore, to better illustrate this application, numerous specific details are provided in the following detailed embodiments. Those skilled in the art should understand that this application can be implemented without certain specific details. In some instances, methods, means, components, and circuits well-known to those skilled in the art have not been described in detail in order to highlight the main points of this application.

[0082] Existing vehicle collision risk detection or vehicle collision probability calculations typically only consider the behavior of the vehicle and surrounding vehicles, without considering collisions between the vehicle's forward-moving vehicles and surrounding vehicles, or between surrounding vehicles themselves. As a result, the accuracy of the obtained vehicle collision probability is poor, and it cannot effectively reduce the incidence of vehicle collision accidents.

[0083] Figure 1 This diagram illustrates a prior art method for calculating collision probability. For example... Figure 1 As shown, this technical solution considers only the vehicle's speed when calculating the vehicle collision probability. Based on the vehicle's speed, the area in front of the vehicle is divided into elliptical regions corresponding to different collision probability levels (e.g., 0%, 20%, 40%, 80%, 100%). The region where the target object (e.g., a pedestrian) is located is extracted from the image captured by the vehicle. The collision probability value is obtained based on the collision probability of the elliptical region in which the target object is located.

[0084] However, this technical solution only determines the collision probability based on an elliptical region defined by the vehicle's speed, without considering the predicted behavior of the vehicle and surrounding vehicles or objects. A vehicle traveling at high speed may have a lower collision probability at one moment, but a higher probability at the next. Therefore, this technical solution cannot predict collision behavior in advance, resulting in the inability to calculate collision probabilities in a timely manner when the vehicle is traveling at high speed, and the accuracy of the calculated collision probabilities is poor, leading to a higher vehicle safety risk.

[0085] However, the predicted driving trajectory based on the vehicle's historical trajectory and current motion state may deviate significantly from the actual future trajectory. For example, Figure 2 This diagram illustrates a situation where the predicted trajectory of a vehicle deviates significantly from its actual future trajectory. Figure 2 As shown, the predicted trajectory L1 of vehicle A is determined based on historical trajectory points and its current motion state. However, during actual driving, if vehicle B merges into the lane currently being driven by vehicle A, the driver of vehicle A will take reasonable evasive action to avoid vehicle B. The actual future trajectory of vehicle A will then be L2, which deviates significantly from the predicted trajectory L1. In this situation, the accuracy of the vehicle collision probability calculated based on the predicted trajectory is low.

[0086] To address the aforementioned technical problems, this application provides a vehicle collision probability detection method. Based on the premise that a rational driver will make reasonable avoidance maneuvers according to the vehicle's environment during driving, the vehicle collision probability detection method of this application considers not only the predicted trajectory but also the planned trajectory for obstacle avoidance during vehicle trajectory calculation. Furthermore, it calculates the collision probability between the current vehicle and each surrounding vehicle, as well as the collision probability among surrounding vehicles, based on the fusion of the predicted trajectory and the trajectory of the current vehicle (i.e., the self-vehicle), and the fusion of the predicted trajectories and trajectories of surrounding vehicles. This improves the accuracy of vehicle collision probability, effectively reducing the incidence of collision accidents and enhancing the driving safety of the self-vehicle.

[0087] The vehicle collision probability detection method of this application embodiment can be applied to in-vehicle terminal equipment. The in-vehicle terminal equipment can be an independent terminal device, which can be connected to control systems such as Adaptive Cruise Control (ACC), Automatic Emergency Braking (AEB), and Advanced Driving Assistance System (ADAS) to provide vehicle collision probability detection for the control system.

[0088] In one possible implementation, the in-vehicle terminal equipment can also be integrated into control systems such as Adaptive Cruise Control (ACC), Automatic Emergency Braking (AEB), and Advanced Driver Assistance Systems (ADAS) to provide vehicle collision probability detection for the control system.

[0089] Figure 3a This diagram illustrates an application scenario of a vehicle collision probability detection method according to an embodiment of this application. Figure 3aAs shown, the vehicle collision probability detection method can be applied to the on-board terminal device 20 of vehicle 10. The on-board terminal device 20 is connected to the control system 30 and provides the control system 30 with vehicle collision probability detection. The control system 30 can be, for example, ACC, AEB, ADAS, etc.

[0090] Figure 3b This diagram illustrates an application scenario of a vehicle collision probability detection method according to an embodiment of this application. Figure 3b As shown, the vehicle collision probability detection method can be applied to the vehicle-mounted terminal device 20 of the vehicle 10. The vehicle-mounted terminal device 20 is integrated into the control system 30 to provide the control system 30 with vehicle collision probability detection.

[0091] The vehicle collision probability detection method of this application embodiment can also be applied to an in-vehicle terminal device and a server. The in-vehicle terminal device and the server are connected via a network, and the in-vehicle terminal device and the server are respectively used to perform different processes in the vehicle collision probability detection method. For example, the in-vehicle terminal device can be used to obtain the status information of the current vehicle and its surrounding vehicles, reference lines (such as the center line of the lane where the vehicle is located), and other information, and send them to the server; the server can be used to calculate the collision probability based on the information obtained by the in-vehicle terminal device, and send the calculation result (i.e., the collision probability) to the in-vehicle terminal device so that the in-vehicle terminal device can perform subsequent processing. It should be understood that those skilled in the art can set the specific processing performed by the in-vehicle terminal device and the server according to the actual situation, and this application does not limit this.

[0092] Figure 3c This diagram illustrates an application scenario of a vehicle collision probability detection method according to an embodiment of this application. Figure 3c As shown, the vehicle collision probability detection method is applied to the vehicle-mounted terminal device 20 and server 40 of vehicle 10. The server 40 and the vehicle-mounted terminal device 20 are connected via a network. The vehicle-mounted terminal device 20 is used to acquire information about the current vehicle and its surrounding vehicles, and sends the acquired information to the server 40 for processing. The server 40 calculates the collision probability based on the received information and sends the calculation result to the vehicle-mounted terminal device 20. The vehicle-mounted terminal device 20 receives the collision probability calculated by the server 40 and sends it to the control system 30 of vehicle 10 for corresponding processing.

[0093] Figure 4 A flowchart illustrating a vehicle collision probability detection method according to an embodiment of this application is shown. Figure 4 As shown, the vehicle collision probability detection method includes:

[0094] Step S401: Obtain the status information and reference line of the current vehicle, and the status information and reference line of at least one target vehicle. The status information includes at least one of lateral position, longitudinal position, speed, acceleration, heading angle and heading angular velocity. The reference line of the current vehicle includes the center line of the lane where the current vehicle is located. The at least one target vehicle is a vehicle surrounding the current vehicle.

[0095] Step S402: Based on the current vehicle status information, predict the current vehicle's driving trajectory;

[0096] Step S403: Based on the status information of the at least one target vehicle, predict the driving trajectory of the at least one target vehicle;

[0097] Step S404: Based on the current vehicle's status information and reference line, and the status information and reference line of the at least one target vehicle, obtain the planned trajectory of the current vehicle and the planned trajectory of the at least one target vehicle, wherein the fusion of the predicted driving trajectory and the planned trajectory indicates the collision probability between the current vehicle and each target vehicle and the collision probability between each target vehicle.

[0098] In one possible implementation, when detecting or calculating the probability of a vehicle collision, in step S401, the current vehicle's (i.e., the vehicle's own vehicle) status information and reference line, as well as the status information and reference line of at least one target vehicle, can be obtained through sensing components such as video cameras, radar sensors, and ultrasonic sensors.

[0099] Among them, at least one target vehicle is a vehicle surrounding the current vehicle. The status information of the current vehicle and each target vehicle may include at least one of the following: lateral position, longitudinal position, speed, acceleration, heading angle, and heading angular velocity.

[0100] When the vehicle's turn signal is not activated, the vehicle's reference line can include the center line of the lane it is currently in. The lane can be determined by the vehicle's center of gravity. In other words, the vehicle's reference line can be the center line of the lane where its center of gravity is located. When the vehicle's turn signal is activated, the vehicle's reference line is the center line of the adjacent lane corresponding to the turn signal.

[0101] The method for determining the reference lines of each target vehicle in at least one target vehicle is similar to the method for determining the reference lines of the current vehicle, and will not be repeated here.

[0102] Figure 5 A schematic diagram showing a vehicle reference line in a vehicle collision probability detection method according to an embodiment of this application is provided. Figure 5As shown in the diagram, there are three lanes with center lines C1, C2, and C3. Vehicle A1 has not activated its turn signal, and the center line of the lane containing vehicle A1's center of gravity is C1; therefore, vehicle A1's reference line can be determined as C1. Vehicle A2 has not activated its turn signal, but the center line of the lane containing vehicle A2's center of gravity is C2; therefore, vehicle A2's reference line can be determined as C2. Vehicle A3 has not activated its turn signal, and the center line of the lane containing vehicle A3's center of gravity is C2; therefore, vehicle A3's reference line can be determined as C2. Vehicle A4 has activated its left turn signal; vehicle A4's reference line is the center line C2 of the adjacent lane corresponding to its left turn signal.

[0103] In one possible implementation, after obtaining the current vehicle's state information and reference line, and the state information and reference line of at least one target vehicle, in step S402, the current vehicle's trajectory can be predicted using various methods such as position prediction and position fitting based on the current vehicle's state information. This application does not limit the specific method used to predict the current vehicle's trajectory based on its state information.

[0104] In one possible implementation, step S402 may include: determining at least one future position of the current vehicle based on the current vehicle's state information (including at least one of lateral position, longitudinal position, speed, acceleration, heading angle, and heading angular velocity); and predicting the current vehicle's trajectory based on the at least one future position.

[0105] For example, when predicting the current vehicle's trajectory, assume the current vehicle's state information is [x0, y0, v0, a0, θ0, ...]. In this equation, x0 represents the lateral position, y0 represents the longitudinal position, v0 represents the velocity, a0 represents the acceleration, and θ0 represents the heading angle. Let x represent the heading angular velocity. Assuming the number of future positions is N (N is a positive integer), the i-th future position (x) of the current vehicle can be determined by the following formula (1). i ,y i ), and the current vehicle's future location (x i ,y i The velocity v at time ) i and heading angle θ i :

[0106] x i =x i-1 +v i-1 *T*cos(θ i-1 );

[0107] y i =y i-1 +v i-1 *T*sin(θ i-1 );

[0108] v i =v0 + a0 * i * T;

[0109]

[0110] In formula (1), i is a positive integer and i≤N, T represents the time interval for position sampling, (x i-1 ,y i-1 ) represents the (i-1)th future position of the current vehicle, v i-1 Indicates the current vehicle's future location (x) i-1 ,y i-1 The velocity at θ i-1 Indicates the current vehicle's future location (x) i-1 ,y i-1 The heading angle at that time.

[0111] The above method determines the N future positions (x1, y1), ..., (x N ,y N After that, a Frenet coordinate system can be established based on the current vehicle's reference line. Then, the N future positions located in the Cartesian coordinate system can be transformed to the Frenet coordinate system, resulting in N future positions (s1, d1), ..., (s...) in the Frenet coordinate system. N ,d N ), where the i-th future position of the current vehicle in the Flyner coordinate system can be represented as (s i ,d i ).

[0112] It should be understood that the N future positions of the current vehicle can be transformed from the Cartesian coordinate system to the Freina coordinate system in various ways. The following is only an example of coordinate transformation, and this application does not limit the specific method of coordinate transformation.

[0113] Figure 6 This diagram illustrates the coordinate transformation of a vehicle collision probability detection method according to an embodiment of this application. During coordinate transformation, the reference line of the current vehicle can be represented as a polynomial: y = C 0r +C 1r *x+C 2r *x 2 +C 3r *x 3 , where C 0r C 1r C 2r C 3r These are polynomial coefficients, which can be obtained and determined through sensing components such as sensors. For example... Figure 6 As shown, in the Cartesian coordinate system, the reference line of the current vehicle is L3, and the current vehicle's trajectory is L4. During coordinate transformation, a perpendicular line can be drawn from the current position (x0, y0) of the current vehicle to the reference line L3 to obtain the foot of the perpendicular point F1, whose coordinates are represented as (x0, y0). ds ,y ds ), from the current vehicle's i-th future position (x i ,y i Draw a perpendicular line to the reference line L3, obtaining the foot of the perpendicular point F2, whose coordinates are represented as (x...). de ,y de Then, the i-th future position (s) of the current vehicle in the Fleischer coordinate system can be calculated using the following formula (2). i ,d i d in ) i :

[0114]

[0115] In formula (2), in [(y i -y de cosθ de -(x i -x de sinθ de When d > 0, i If positive, otherwise d i Negative,

[0116] The i-th future position (s) of the current vehicle in the FRENER coordinate system can be calculated using the following formula (3). i ,d i s in ) i :

[0117]

[0118] Using the above method, we can obtain N future positions (s1, d1), ..., (s...) of the current vehicle in the Fleischer coordinate system. N ,d N Then, the N future positions (s1, d1), ..., (s...) of the current vehicle can be determined using methods such as least squares. N ,d N The vehicle's predicted trajectory is obtained by fitting the data. The predicted trajectory can be represented by the following formula (4):

[0119]

[0120] In formula (4), s pre (t)=f s(t) represents the longitudinal displacement s with respect to time t, d pre (t)=f d (t) is a curve representing the lateral displacement d with respect to time t.

[0121] Figure 7 Figure 7 illustrates a schematic diagram of the predicted driving trajectory of a vehicle collision probability detection method according to an embodiment of this application. As shown in Figure 7, the least squares method can be used to predict the N future positions (s1, d1), ..., (s...) of the current vehicle. N ,d N The system is fitted to obtain the predicted driving trajectory L5 of the current vehicle.

[0122] In this way, the trajectory of a vehicle can be predicted based on at least one future location, which is simple and fast, thereby improving processing efficiency.

[0123] In one possible implementation, step S402 may include: obtaining at least one historical location of the current vehicle; determining at least one future location of the current vehicle based on the current vehicle's status information; and predicting the current vehicle's driving trajectory based on the at least one historical location and the at least one future location.

[0124] In other words, the current vehicle's trajectory can also be predicted based on at least one historical location and at least one future location of the current vehicle. The method for obtaining at least one future location of the current vehicle in this implementation is similar to the method described above, and will not be repeated here.

[0125] When acquiring at least one historical position of the current vehicle, M′ historical positions of the current vehicle (M′ is a positive integer) can be recorded sequentially according to the position sampling time interval T. When recording historical positions, the historical position (x) of the current vehicle at the previous time step must be included. j ,y j ), where j is a positive integer and j≤M′, based on the speed v of the vehicle at the current time in the previous time step. j and angular velocity of heading Perform coordinate rotation and translation to ensure that at least one historical position is represented in the current coordinate system. The coordinate rotation amount is... The coordinate translation is v j T.

[0126] Figure 8 This diagram illustrates the coordinate rotation and translation of a vehicle collision probability detection method according to an embodiment of this application. Figure 8As shown, 82 represents the current vehicle's position information at the current moment, including one current position (represented by a white dot) and three historical positions (represented by black dots). 81 represents the current vehicle's position information at the previous moment, including one current position (represented by a white dot) and two historical positions (represented by black dots). From the previous moment to the current moment, the coordinates of the three historical positions have been rotated and translated.

[0127] After obtaining at least one historical position and at least one future position of the current vehicle, the positions in the Cartesian coordinate system can be transformed to the Flexner coordinate system. Then, the least historical position and at least one future position in the Flexner coordinate system are fitted using the least squares method to obtain the predicted driving trajectory of the current vehicle. The specific processing methods for coordinate transformation and position fitting are similar to those described above and will not be repeated here.

[0128] In this way, the current vehicle's trajectory can be predicted based on at least one historical location and at least one future location, thereby improving the accuracy of the predicted trajectory.

[0129] In one possible implementation, when predicting the trajectory of the current vehicle based on at least one future location, or based on at least one historical location and at least one future location, the current location of the current vehicle may also be considered. For example, the trajectory of the current vehicle may be predicted based on its current location and at least one future location. Alternatively, the trajectory of the current vehicle may be predicted based on its current location, at least one historical location, and at least one future location, thereby considering its current location when predicting the trajectory and improving the accuracy of the predicted trajectory.

[0130] In one possible implementation, in step S403, the driving trajectory of the at least one target vehicle can be predicted based on the state information of the at least one target vehicle. Each of the at least one target vehicle has a predicted driving trajectory. The driving trajectories of each target vehicle can be predicted using a method similar to that used to predict the driving trajectory of the current vehicle, which will not be elaborated here.

[0131] In one possible implementation, in step S404, the planned trajectory of the current vehicle and the planned trajectory of at least one target vehicle are obtained based on the current vehicle's state information and reference lines, and the state information and reference lines of at least one target vehicle. The planned trajectory can be used to represent the possible driving trajectory of the vehicle when avoiding surrounding obstacles.

[0132] For the current vehicle, obstacle information around the current vehicle can be determined based on its status information and reference lines, as well as the status information and reference lines of at least one target vehicle. Then, the planned trajectory of the current vehicle can be determined based on the obstacle information, the current vehicle's status information, and reference lines.

[0133] For any target vehicle, the obstacle information around the target vehicle can be determined based on the target vehicle's status information and reference lines, as well as the status information and reference lines of the surrounding vehicles. Then, based on the obstacle information, the target vehicle's status information, and reference lines, the planned trajectory of the target vehicle can be determined.

[0134] In one possible implementation, step S404 may include: for the current vehicle, determining at least one planned position of the current vehicle based on the current vehicle's status information and reference line, the status information and reference line of at least one target vehicle, and preset position intervals and speed intervals; and determining the planned trajectory of the current vehicle based on the at least one planned position.

[0135] The preset position interval is the lateral displacement interval in the Flener coordinate system, and the preset velocity interval is the longitudinal velocity interval in the Flener coordinate system.

[0136] When determining the planned trajectory of the current vehicle, obstacle information around the current vehicle can be determined based on the current vehicle's status information and reference lines, as well as the status information and reference lines of at least one target vehicle. Then, based on the obstacle information, the current vehicle's status information and reference lines, and preset position and speed intervals, at least one planned position for the current vehicle can be determined. Each planned position of the current vehicle takes into account the avoidance of potential obstacles. The planned trajectory of the current vehicle can be determined based on at least one planned position. Each planned position can correspond to one planned trajectory.

[0137] In one possible implementation, the current vehicle has at least one planned trajectory. The starting point of each planned trajectory is the current position of the current vehicle, and the ending point of each planned trajectory corresponds to a planned position.

[0138] Figure 9 This diagram illustrates the path planning of a vehicle collision probability detection method according to an embodiment of this application. Figure 9 As shown, the current vehicle's reference line is L6. Preset position and speed intervals can be represented by a grid. Based on the current vehicle's status information and the reference line, as well as the status information and reference line of at least one target vehicle, obstacle information around the current vehicle can be determined. While avoiding obstacles, at least one planned position for the current vehicle can be determined using the grid. Figure 9Nineteen planned locations were identified, and then, starting from the current location of the vehicle, each planned location was used as the end point to determine the 19 planned trajectories of the vehicle.

[0139] In one possible implementation, each planned trajectory can be represented by a polynomial. Those skilled in the art can set the degree of the polynomial according to the actual situation; this application does not impose any restrictions on this.

[0140] In this way, based on the current vehicle's status information and reference line, the status information and reference line of at least one target vehicle, and preset position and speed intervals, at least one planned position of the current vehicle can be determined, and based on at least one planned position, the planned trajectory of the current vehicle can be determined, thereby improving the accuracy of the planned trajectory of the current vehicle.

[0141] In one possible implementation, the planned trajectories of each of the target vehicles in at least one target vehicle can be determined in a manner similar to determining the planned trajectory of the current vehicle, which will not be elaborated here.

[0142] In one possible implementation, after obtaining the predicted driving trajectory and planned trajectory of the current vehicle, and the predicted driving trajectory and planned trajectory of each of the at least one target vehicle, the collision probability between the current vehicle and each target vehicle, as well as the collision probability between each target vehicle, can be indicated by fusing the predicted driving trajectory and the planned trajectory.

[0143] According to embodiments of this application, it is possible to obtain the status information and reference lines of the current vehicle (self-vehicle), and the status information and reference lines of at least one target vehicle (vehicles surrounding the self-vehicle). Based on the status information of the current vehicle, the driving trajectory of the current vehicle is predicted. Based on the status information of at least one target vehicle, the driving trajectory of at least one target vehicle is predicted. Based on the status information and reference lines of the current vehicle and at least one target vehicle, the planned trajectory of the current vehicle and the planned trajectory of at least one target vehicle are obtained. The fusion of the predicted driving trajectory and the planned trajectory indicates the collision probability between the current vehicle and each target vehicle, as well as the collision probability between each target vehicle. This can improve the accuracy of vehicle collision probability detection, thereby effectively reducing the incidence of vehicle collision accidents and improving the driving safety of the self-vehicle.

[0144] Figure 10 A flowchart illustrating a vehicle collision probability detection method according to an embodiment of this application is shown. Figure 10As shown, the vehicle collision probability detection method of this embodiment may include steps S401, S402, S403, S404, S405, S406, and S407. Steps S401, S402, S403, and S404 are similar to those in the above embodiment and will not be described again here.

[0145] In this embodiment, the predicted driving trajectory and planned trajectory of the current vehicle, as well as the predicted driving trajectory and planned trajectory of each of the at least one target vehicle, can be determined through steps S401, S402, S403, and S404. Then, in step S405, the predicted driving trajectory and planned trajectory of the current vehicle can be fused to obtain the fused trajectory of the current vehicle. For example, point fusion, coefficient fusion, etc. can be used, and trajectory fusion can also adopt other methods, which are not limited in this application.

[0146] In one possible implementation, the current vehicle has a predicted driving trajectory and at least one planned trajectory. When performing trajectory fusion, the predicted driving trajectory of the current vehicle can be fused with each of the current vehicle's planned trajectories to obtain at least one fused trajectory for the current vehicle.

[0147] Figure 11 A schematic diagram illustrating trajectory fusion in a vehicle collision probability detection method according to an embodiment of this application is shown. Figure 11 As shown, L7 represents the predicted driving trajectory of the current vehicle, L8 represents any planned trajectory of the current vehicle, and L9 represents the fused trajectory of the current vehicle.

[0148] In one possible implementation, step S405 may include: discretizing the predicted driving trajectory and the planned trajectory of the current vehicle according to a preset second time series to obtain multiple predicted trajectory points and multiple planned trajectory points; and determining the fused trajectory of the current vehicle according to a preset weight sequence, the multiple predicted trajectory points and the multiple planned trajectory points.

[0149] The second time series includes multiple discrete moments with the same interval length, and the preset weight series includes multiple weight values. As time increases, the weight of the predicted driving trajectory gradually decreases, while the weight of the planned trajectory gradually increases.

[0150] Assuming the second time series comprises K discrete moments with equal intervals (where K is a positive integer), it can be represented by a function T(t) of time. Correspondingly, the weight sequence also comprises K weight values, which can be represented as W(t). During trajectory fusion, the predicted trajectory of the current vehicle is discretized based on the second time series, yielding K predicted trajectory points. Similarly, any planned trajectory of the current vehicle is discretized based on the second time series, yielding K planned trajectory points. Then, based on the weight sequence, the K predicted trajectory points, and the K planned trajectory points, K fusion trajectory points are determined. Finally, the K fusion trajectory points are fitted using the least squares method to obtain the fusion trajectory of the current vehicle.

[0151] In one possible implementation, the fused trajectory T of the current vehicle can be determined by the following formula (5). final :

[0152] T final =[T pre *W(t)+T plan *(1-W(t))] (5)

[0153] In formula (5), T pre T represents the predicted driving trajectory of the current vehicle. plan Let W(t) represent any planned trajectory of the current vehicle, and let W(t) represent the weight, which is a function of time t. W(t) decreases as t increases, thus making T... pre The weight of T gradually decreases as t increases. plan The weight of the fusion trajectory T gradually increases with the increase of t, thereby improving the fusion trajectory T. final The accuracy.

[0154] When performing trajectory fusion using the above formula (5), the second time series T(t) can be used to calculate T(t) respectively. pre and T plan Discretization is performed to obtain K predicted trajectory points {s}. pre (t),d pre (t)}, K planned trajectory points {s plan (t), d plan (t)}, and according to the weight W(t), the corresponding trajectory points in the K predicted trajectory points and the K planned trajectory points are successively fused to obtain K fused trajectory points {s final (t), d final (t)}, where s final (t)=s pre (t)*W(t)+s plan (t)*(1-W(t)), d final (t)=d pre(t)*W(t)+d plan (t)*(1-W(t)), then use the least squares method to fuse the K trajectory points {s final (t), d final Fitting the fusion trajectory T to obtain the fusion trajectory T final .

[0155] By using point fusion, the predicted driving trajectory and the planned trajectory of the current vehicle are merged to obtain the fused trajectory of the current vehicle, which can improve the accuracy of the fused trajectory.

[0156] In one possible implementation, step S405 may include: determining a first polynomial coefficient corresponding to the predicted driving trajectory of the current vehicle and a second polynomial coefficient corresponding to the planned trajectory of the current vehicle; and determining the fused trajectory of the current vehicle based on preset coefficient weights, the first polynomial coefficient, and the second polynomial coefficient.

[0157] When determining the fusion trajectory of the current vehicle through coefficient fusion, the predicted driving trajectory of the current vehicle can be represented as a first polynomial, and the planned trajectory of the current vehicle can be represented as a second polynomial, wherein the degree of the second polynomial is the same as that of the first polynomial. This yields the coefficients of the first polynomial corresponding to the predicted driving trajectory of the current vehicle and the coefficients of the second polynomial corresponding to the planned trajectory of the current vehicle. Then, based on the preset coefficient weights, the coefficients of the first polynomial, and the coefficients of the second polynomial, the polynomial coefficients of the fusion trajectory of the current vehicle are determined, and the fusion trajectory of the current vehicle is obtained based on the polynomial coefficients of the fusion trajectory.

[0158] For example, the first polynomial of the predicted trajectory of the current vehicle is represented as follows:

[0159]

[0160]

[0161] in, The coefficients of the first polynomial.

[0162] The second polynomial of any planned trajectory of the current vehicle is represented as follows:

[0163]

[0164]

[0165] in, These are the coefficients of the second polynomial.

[0166] The u-th coefficient of the polynomial of the fused trajectory of the current vehicle can be determined in the following way:

[0167]

[0168]

[0169] Where u takes values ​​of 0, 1, 2, and 3, W a (u) represents the coefficient weights of the first polynomial, W b (u) represents the coefficient weights of the second polynomial. Denotes the u-th coefficient of the first polynomial. Let u represent the u-th coefficient of the second polynomial.

[0170] The fused trajectory of the current vehicle can be obtained based on the polynomial coefficients of the fused trajectory. The fused trajectory can be represented as follows:

[0171]

[0172]

[0173] By using coefficient fusion, the predicted driving trajectory and the planned trajectory of the current vehicle are fused to obtain the fused trajectory of the current vehicle. This not only improves the accuracy of the fused trajectory, but also reduces the amount of computation and improves processing efficiency.

[0174] In one possible implementation, in step S406, the predicted driving trajectory of the at least one target vehicle is fused with the planned trajectory to obtain the fused trajectory of the at least one target vehicle.

[0175] Similar to the current vehicle, any target vehicle has a predicted driving trajectory and at least one planned trajectory. During trajectory fusion, the predicted driving trajectory of the target vehicle can be fused with each of its planned trajectories to obtain at least one fused trajectory for the target vehicle. The specific method of trajectory fusion is similar to that of the current vehicle's trajectory fusion, and will not be elaborated here.

[0176] In one possible implementation, after determining the fusion trajectory of the current vehicle and at least one target vehicle, in step S407, based on the fusion trajectory of the current vehicle and the fusion trajectory of the at least one target vehicle, the collision probability between the current vehicle and each target vehicle, as well as the collision probability between each target vehicle, can be determined through various methods such as trajectory overlap, region overlap, and trajectory cost.

[0177] In one possible implementation, step S407 may include: determining a trajectory set of a first vehicle, wherein the first vehicle is any one of the current vehicle and the at least one target vehicle, the trajectory set includes at least one trajectory combination, each trajectory combination including a fused trajectory of the first vehicle and a fused trajectory of a second vehicle, wherein the second vehicle is any one of the current vehicle and the at least one target vehicle other than the first vehicle; determining the cost value of each trajectory combination in the trajectory set; and determining the collision probability between the first vehicle and the second vehicle based on the cost value of each trajectory combination in the trajectory set.

[0178] In one possible implementation, any one of the current vehicle and at least one target vehicle can be considered as the first vehicle, and any vehicle other than the first vehicle among the current vehicle and at least one target vehicle can be considered as the second vehicle. When determining the collision probability of the first vehicle, the fused trajectory of the first vehicle can be combined with the fused trajectories of each of the second vehicles to obtain a trajectory set of the first vehicle. This trajectory set includes at least one trajectory combination, and each trajectory combination includes a fused trajectory of the first vehicle and a fused trajectory of the second vehicle.

[0179] For example, suppose the fused trajectory of the first vehicle is T. target ={t0, ...,t} p}, where p is a positive integer representing the number of fused trajectories of the first vehicle. Assume the number of target vehicles is R, the number of fused trajectories for any target vehicle is m, and the fused trajectories of R target vehicles are... Where R and m are positive integers, T target With T other The fused trajectories are combined pairwise to obtain the trajectory set T of the first vehicle. comb ={(t0,o0),(t0,o1),…,(t p-1 ,o q-1 )}, where q=R×m.

[0180] After determining the trajectory set of the first vehicle, the cost value of each trajectory combination within that set can be determined. Based on comfort and target deviation, the cost value of each trajectory combination can be determined according to at least one of the following: longitudinal acceleration, longitudinal velocity deviation, lateral acceleration, and lateral position deviation. A higher cost value for a trajectory combination indicates that the fused trajectory within that combination deviates more from the driving habits of a rational driver.

[0181] The collision probability between the first vehicle and the second vehicle can be determined based on the cost values ​​of each trajectory combination in the trajectory set of the first vehicle. For example, the minimum cost value (i.e., the minimum value among the cost values) can be determined from the cost values ​​of each trajectory combination in the trajectory set by comparison, sorting, and selecting the minimum value. Based on this minimum cost value, the collision probability between the first vehicle and the second vehicle can be determined through a preset transformation relationship, such as a transformation table or transformation function. The larger the cost value, the greater the corresponding collision probability.

[0182] In this way, a trajectory set of the first vehicle can be established and the cost value of each trajectory combination in the trajectory set can be determined. Based on the cost value of each trajectory combination in the trajectory set, the collision probability between the first vehicle and the second vehicle can be determined, which can improve the accuracy of the collision probability.

[0183] In one possible implementation, after determining the trajectory set of the first vehicle, it is also possible to determine whether a collision occurs between the various trajectory combinations in the trajectory set of the first vehicle through methods such as trajectory overlap and region overlap.

[0184] For example, for any combination of trajectories in the trajectory set of the first vehicle, the fused trajectory of the first vehicle in the trajectory combination can be discretized according to a preset first time series including multiple times to obtain multiple target positions of the first vehicle, and the fused trajectory of the second vehicle in the trajectory combination can be discretized to obtain multiple target positions of the second vehicle.

[0185] For any given moment in the first time series, at that moment, the target area where the first vehicle is located and the target area where the second vehicle is located are determined. The target area where the first vehicle is located corresponds to the target position of the first vehicle at that moment, and the target area where the second vehicle is located corresponds to the target position of the second vehicle at that moment. The size of the target area is larger than the vehicle body size, meaning the target area is large enough to cover the vehicle body with sufficient safety margins in both the longitudinal and lateral directions. The target area can be a rectangular area or a three-circle area, which allows for a higher degree of conformity between the target area and the vehicle's shape. The three-circle area includes three circular areas defined by the center of the front axle, the center of the vehicle, and the center of the rear axle, respectively, with the wheelbase of the vehicle as the diameter. Currently, the area can also be other shapes, and this application does not impose any restrictions on this.

[0186] After determining the target area, it can be judged whether the target area where the first vehicle is located overlaps with the target area where the second vehicle is located. If the target areas of the first vehicle and the second vehicle overlap, it indicates that the trajectory combination will collide; if the target areas of the first vehicle and the second vehicle do not overlap, it indicates that the trajectory combination will not collide. Judging whether a collision will occur by using area overlap can improve the accuracy of collision detection for trajectory combinations.

[0187] Figure 12 A schematic diagram showing the target area of ​​a vehicle collision probability detection method according to an embodiment of this application is provided. Figure 12 As shown, at each time point, the target areas where vehicles B1 and B2 are located are three-circle areas. For any given time point, the overlap between the target areas of vehicle B1 and vehicle B2 is determined by whether the three circles of the two vehicles overlap.

[0188] Figure 13 A schematic diagram showing the target area of ​​a vehicle collision probability detection method according to an embodiment of this application is provided. Figure 13 As shown, the target areas of vehicles B1 and B2 are rectangular regions. Assuming that the top left vertex of the target area of ​​vehicle B1 is (x1, y1) and the bottom right vertex is (x2, y2), and the top left vertex of the target area of ​​vehicle B2 is (x3, y3) and the bottom right vertex is (x4, y4), if (x2≤x3)||(y2≤y3)||(x1≥x4)||(y1≥y4), then the target areas of vehicles B1 and B2 are considered to have no overlap; otherwise, the target areas of vehicles B1 and B2 are considered to have overlap.

[0189] Based on the judgment results of whether a collision occurs in each trajectory combination in the trajectory set of the first vehicle, trajectory combinations that do not collide can be selected from the trajectory combinations of the first vehicle to establish a non-collision trajectory set, and the cost value of each trajectory combination in the non-collision trajectory set can be determined. Among them, trajectory combinations that collide are unsafe trajectory combinations with a collision probability of 1; trajectory combinations that do not collide are potentially unsafe trajectory combinations, and their collision probability can be determined by the cost value.

[0190] In one possible implementation, the cost C of any combination of trajectories can be determined by the following formula (6). gh :

[0191]

[0192] Where g and h are integers, and 0 ≤ g ≤ m, 0 ≤ h ≤ q, K′ represents the number of moments in the time series when discretizing the fused trajectory in the trajectory combination, P = 0 represents the current moment, and g represents the fused trajectory T of the current vehicle. target In the t, the g-th and h-th represent the fused trajectories T of the R target vehicles. other The h-th one;

[0193] For the g-th fusion trajectory d g The third derivative of (t) represents the lateral jerk of the current vehicle; For the g-th fusion trajectory sg The third derivative of (t) represents the current longitudinal acceleration of the vehicle; d g (K′) represents the lateral position of the current vehicle on the g-th fused trajectory at time K′; This represents the target value of the lateral position of the current vehicle on the g-th fused trajectory; This represents the longitudinal velocity of the current vehicle on the g-th fused trajectory at time K′. This represents the target value of the longitudinal velocity of the current vehicle on the g-th fused trajectory;

[0194] For the h-th fusion trajectory d h The third derivative of (t) represents the lateral jerk of the target vehicle corresponding to the h-th fused trajectory; For the h-th fusion trajectory s h The third derivative of (t) represents the longitudinal jerk of the target vehicle corresponding to the h-th fused trajectory; d h (K′) represents the lateral position of the target vehicle corresponding to the h-th fused trajectory at time K′; The target value represents the lateral position of the target vehicle corresponding to the h-th fused trajectory on the h-th fused trajectory; This represents the longitudinal velocity of the target vehicle corresponding to the h-th fused trajectory at time K′; This represents the target value of the longitudinal velocity of the target vehicle corresponding to the h-th fused trajectory on the h-th fused trajectory.

[0195] After determining the cost value of each trajectory combination in the set of non-collision trajectories using the above method, the collision probability between the first and second vehicles can be determined based on the cost values ​​of each trajectory combination in the set of non-collision trajectories. For example, the minimum cost value (i.e., the minimum value among the cost values) can be determined from the set of non-collision trajectories by comparison and selection of the minimum value, i.e., C. min =min(C gh Based on this minimum cost, and through a pre-defined conversion relationship, such as a conversion table or conversion function CP = f(C), the conversion is performed. min The collision probability between the first and second vehicles is determined using methods such as […]. The minimum cost C is […]. min The larger the value, the greater the probability of collision (CP) between the first and second vehicles.

[0196] In one possible implementation, the collision probability CP between the first vehicle and the second vehicle can be set to 1 if the set of non-collision trajectories is empty.

[0197] By determining the collision, a set of non-collision trajectories is selected from the trajectory set of the first vehicle to establish a set of non-collision trajectories. The cost value of each trajectory combination in the set of non-collision trajectories is determined. Then, based on the cost value of each trajectory combination in the set of non-collision trajectories, the collision probability between the first vehicle and the second vehicle is determined. This reduces the number of trajectory combinations whose cost values ​​are calculated, thereby improving processing efficiency.

[0198] According to embodiments of this application, a fused trajectory of the current vehicle is obtained by fusing the predicted driving trajectory of the current vehicle with the planned trajectory, and a fused trajectory of at least one target vehicle is obtained by fusing the predicted driving trajectory of at least one target vehicle with the planned trajectory. Based on the fused trajectory of the current vehicle and the fused trajectory of at least one target vehicle, the collision probability between the current vehicle and each target vehicle and the collision probability between each target vehicle are determined. This allows the fusion of the predicted driving trajectory with the planned trajectory that takes into account obstacle avoidance, resulting in a fused trajectory that is more in line with the behavior habits of rational drivers. The collision probability between the vehicle and surrounding vehicles and the collision probability between each surrounding vehicle are determined based on this fused trajectory. This makes the collision probability calculation consider not only the behavior of the vehicle and surrounding vehicles, but also the behavior of surrounding vehicles with each other, thereby improving the accuracy of vehicle collision probability detection, effectively reducing the incidence of vehicle collision accidents, and improving the driving safety of the vehicle.

[0199] In one possible implementation, the method may further include: outputting prompt information, the prompt information including at least one of voice prompt information and visual prompt information, the prompt information being used to indicate the collision probability.

[0200] Voice prompts can be output at a certain frequency or continuously to keep the driver alert. Visual prompts can be displayed on screens in the vehicle's cabin (such as the central control screen or the display screen of the in-vehicle terminal device), such as displaying vehicles that may be involved in a collision, so that the driver can intuitively and quickly understand the vehicles that may be involved in a collision.

[0201] In one possible implementation, when determining the prompt information for the current vehicle, the group of vehicles with the highest collision probability can be selected from the collision probabilities between the current vehicle and each target vehicle, and the collision probabilities between each target vehicle. The collision probability level of the selected group of vehicles is then determined. Based on the type of the selected group of vehicles (current vehicle and target vehicle, or two target vehicles) and the collision probability level, the prompt information is output in the cabin of the current vehicle.

[0202] The collision probability can be categorized into multiple levels, determined by preset collision probability thresholds. For example, preset collision probability thresholds may include a first collision probability threshold, a second collision probability threshold, and a third collision probability threshold, where the first collision probability threshold < the second collision probability threshold < the third collision probability threshold. The collision probability levels are low, medium, and high.

[0203] When the collision probability is greater than or equal to the first collision probability threshold and less than the second collision probability threshold, the collision probability level can be determined as low; when the collision probability is greater than or equal to the second collision probability threshold and less than the third collision probability threshold, the collision probability level can be determined as medium; when the collision probability is greater than or equal to the third collision probability threshold, the collision probability level can be determined as high.

[0204] When the selected group of vehicles with the highest collision probability is the current vehicle and the target vehicle (i.e., the vehicle itself and surrounding vehicles), and the collision probability level is low, the prompt information output in the cabin of the current vehicle includes: a voice prompt at a first frequency, and / or displaying the target vehicle in a first color on the display screen in the cabin.

[0205] When the collision probability level is medium, the prompt information output in the current vehicle's cabin includes: giving a voice prompt at a second frequency, and / or displaying the target vehicle in a second color on the cabin's display screen, wherein the second frequency is greater than the first frequency, and the second color is more conspicuous than the first color, and the conspicuousness of the color can be achieved by making the color brighter or darker.

[0206] When the collision probability level is high, the prompt information output in the current vehicle's cabin includes: continuous voice prompts, and / or displaying the target vehicle in a third color on the cabin's display screen, wherein the third color is more conspicuous than the second color.

[0207] When the selected group of vehicles with the highest collision probability consists of two target vehicles (i.e., surrounding vehicles and surrounding vehicles), and the collision probability level is low, the prompt information output in the current vehicle's cabin includes: displaying the two target vehicles in the first color on the display screen of the current vehicle's cabin.

[0208] When the collision probability level is medium, the prompt information output in the current vehicle's cabin includes: providing a voice prompt at a first frequency, and / or displaying two target vehicles in a second color on the cabin's display screen, wherein the second color is more conspicuous than the first color;

[0209] When the collision probability level is high, the prompt information output in the current vehicle's cabin includes: a voice prompt at a second frequency, and / or displaying two target vehicles in a third color on the cabin's display screen, wherein the second frequency is greater than the first frequency, and the third color is more conspicuous than the second color.

[0210] The above is merely an example to illustrate the prompts displayed in the cabin of the current vehicle. Those skilled in the art can set the content, prompting method, display method, etc. of the prompts according to the actual situation, and this application does not impose any restrictions on this.

[0211] According to embodiments of this application, voice, visual, and other prompts can be output within the vehicle's cabin based on the collision probability, thereby increasing the driver's alertness and improving the vehicle's driving safety.

[0212] In one possible implementation, the method may further include: performing corresponding driving controls on the current vehicle based on the collision probability. The driving controls performed on the current vehicle may differ depending on the level of the collision probability.

[0213] When performing driving control on the current vehicle, the group of vehicles with the highest collision probability can be selected from the collision probability between the current vehicle and each target vehicle, and the collision probability between each target vehicle. The collision probability level of the selected group of vehicles is determined. Then, according to the type of the selected group of vehicles (current vehicle and target vehicle, or two target vehicles) and the collision probability level, the corresponding driving control is performed on the current vehicle.

[0214] For example, if the selected group of vehicles with the highest collision probability is the current vehicle and the target vehicle (i.e., the vehicle itself and surrounding vehicles), the driving control measures performed on the current vehicle when the collision probability level is low include: controlling the current vehicle to begin deceleration; when the collision probability level is medium, the driving control measures performed on the current vehicle include: controlling the current vehicle to accelerate or decelerate, and / or initiating pre-filling preparation for emergency braking; when the collision probability level is high, the driving control measures performed on the current vehicle include: controlling the current vehicle to activate emergency braking.

[0215] When the selected group of vehicles with the highest collision probability consists of two target vehicles (i.e., surrounding vehicles and surrounding vehicles), the driving controls performed on the current vehicle at a low collision probability level include: changing the target vehicle to follow; at a medium collision probability level, the driving controls performed on the current vehicle include: increasing the following distance; and at a high collision probability level, the driving controls performed on the current vehicle include: controlling the current vehicle to begin decelerating, and / or initiating pre-filling preparation for emergency braking.

[0216] The above examples merely illustrate the driving control performed on the current vehicle under different collision probabilities. Those skilled in the art can set specific driving controls for the current vehicle under different collision probabilities according to actual conditions, and this application does not impose any restrictions on this.

[0217] According to the embodiments of this application, corresponding driving control can be performed on the current vehicle based on the collision probability, which can not only reduce the occurrence rate of vehicle collision accidents, but also improve the driving safety of the vehicle.

[0218] In the embodiments of this application, the predicted driving trajectory of a vehicle is fused with its planned trajectory to correct the predicted driving trajectory and obtain a fused trajectory that is more in line with the driving habits of a reasonable driver. Based on the fused trajectory, the probability of collision between the vehicle and surrounding vehicles, and between surrounding vehicles and their surroundings, can be detected, which can improve the accuracy of vehicle collision probability detection, thereby effectively reducing the incidence of vehicle collision accidents and improving the driving safety of the vehicle.

[0219] In one possible implementation, the vehicle collision probabilities determined in the embodiments of this application can also be applied to Intelligent Vehicle Infrastructure Cooperative Systems (IVICS). The IVICS can determine a road segment distribution map based on the vehicle collision probabilities determined in the embodiments of this application. In the road segment distribution map, road segments with high vehicle collision probabilities can be designated as dangerous road segments, and navigation systems can avoid these dangerous road segments or remind drivers on these dangerous road segments to drive cautiously. If a road segment is designated as a dangerous road segment frequently, it can be suggested that road management or relevant government departments strengthen corresponding measures to improve safety.

[0220] Figure 14 A block diagram of a vehicle collision probability detection device according to an embodiment of this application is shown. Figure 14 As shown, the vehicle collision probability detection device can be used in the vehicle terminal device 20. The vehicle collision probability detection device 1400 includes an acquisition unit 1410 and a processing unit 1420.

[0221] The acquisition unit 1410 is used to acquire the status information and reference line of the current vehicle, and the status information and reference line of at least one target vehicle. The status information includes at least one of lateral position, longitudinal position, speed, acceleration, heading angle and heading angular velocity. The reference line of the current vehicle includes the center line of the lane where the current vehicle is located. The at least one target vehicle is a vehicle surrounding the current vehicle.

[0222] The processing unit 1420 is configured to: predict the driving trajectory of the current vehicle based on the current vehicle's state information; predict the driving trajectory of the at least one target vehicle based on the state information of the at least one target vehicle; and obtain the planned trajectory of the current vehicle and the planned trajectory of the at least one target vehicle based on the current vehicle's state information and reference lines, and the at least one target vehicle's state information and reference lines, wherein the fusion of the predicted driving trajectory and the planned trajectory indicates the collision probability between the current vehicle and each target vehicle, and the collision probability between each target vehicle.

[0223] In one possible implementation, the acquisition unit 1410 can acquire the current vehicle's status information and reference line, and the status information and reference line of at least one target vehicle, from a sensing component or a communication component (including but not limited to a camera, radar sensor, ultrasonic sensor, inertial measurement unit, V2X communication unit, map unit, global positioning system GPS, global navigation satellite system GNSS, etc.).

[0224] In one possible implementation, the processing unit 1420 can be a general-purpose processor, a special-purpose processor, or other integrated circuits, microcontrollers, microprocessors, etc., capable of data processing. This application does not limit the hardware implementation of the processing unit.

[0225] In one possible implementation, the processing unit 1420 is further configured to: fuse the predicted driving trajectory of the current vehicle with the planned trajectory to obtain the fused trajectory of the current vehicle; fuse the predicted driving trajectory of the at least one target vehicle with the planned trajectory to obtain the fused trajectory of the at least one target vehicle; and determine the collision probability between the current vehicle and each target vehicle and the collision probability between each target vehicle based on the fused trajectory of the current vehicle and the fused trajectory of the at least one target vehicle.

[0226] In one possible implementation, the device further includes: an output unit for outputting prompt information, the prompt information including at least one of voice prompt information and visual prompt information, the prompt information being used to indicate the collision probability.

[0227] In one possible implementation, the device further includes a control unit for performing corresponding driving control on the current vehicle based on the collision probability.

[0228] In one possible implementation, determining the collision probabilities between the current vehicle and each of the target vehicles, and among the target vehicles themselves, based on the fused trajectory of the current vehicle and the fused trajectory of the at least one target vehicle, includes: determining a trajectory set of a first vehicle, wherein the first vehicle is any one of the current vehicle and the at least one target vehicle, the trajectory set includes at least one trajectory combination, each trajectory combination including the fused trajectory of the first vehicle and the fused trajectory of a second vehicle, wherein the second vehicle is any one of the current vehicle and the at least one target vehicle other than the first vehicle; determining the cost value of each trajectory combination in the trajectory set; and determining the collision probability between the first vehicle and the second vehicle based on the cost value of each trajectory combination in the trajectory set.

[0229] In one possible implementation, determining the cost value of each trajectory combination in the trajectory set includes: determining whether each trajectory combination in the trajectory set collides; identifying the trajectory combinations that do not collide as a non-collision trajectory set; and determining the cost value of each trajectory combination in the non-collision trajectory set.

[0230] Determining the collision probability between the first vehicle and the second vehicle based on the cost value of each trajectory combination in the trajectory set includes: determining the collision probability between the first vehicle and the second vehicle based on the cost value of each trajectory combination in the non-collision trajectory set.

[0231] In one possible implementation, determining whether a collision occurs between the various trajectory combinations in the trajectory set includes: for any trajectory combination, discretizing the fused trajectory of the first vehicle and the fused trajectory of the second vehicle in the trajectory combination according to a preset first time series to obtain multiple target positions of the first vehicle and multiple target positions of the second vehicle, wherein the first time series includes multiple time points; at any time point in the first time series, determining whether the target area where the first vehicle is located overlaps with the target area where the second vehicle is located, wherein the target area where the first vehicle is located corresponds to the target position of the first vehicle at that time, and the target area where the second vehicle is located corresponds to the target position of the second vehicle at that time, wherein the overlap of the target areas indicates that a collision has occurred between the trajectory combinations.

[0232] In one possible implementation, the target area is a rectangular area or a three-circle area, wherein the three-circle area includes three circular areas defined by the center of the front axle, the center of the vehicle, and the center of the rear axle, respectively, and the wheelbase of the vehicle as the diameter.

[0233] In one possible implementation, fusing the predicted driving trajectory and the planned trajectory of the current vehicle to obtain the fused trajectory of the current vehicle includes: discretizing the predicted driving trajectory and the planned trajectory of the current vehicle according to a preset second time series to obtain multiple predicted trajectory points and multiple planned trajectory points; and determining the fused trajectory of the current vehicle according to a preset weight sequence, the multiple predicted trajectory points and the multiple planned trajectory points.

[0234] In one possible implementation, fusing the predicted driving trajectory and the planned trajectory of the current vehicle to obtain the fused trajectory of the current vehicle includes: determining a first polynomial coefficient corresponding to the predicted driving trajectory of the current vehicle and a second polynomial coefficient corresponding to the planned trajectory of the current vehicle; and determining the fused trajectory of the current vehicle according to preset coefficient weights, the first polynomial coefficient, and the second polynomial coefficient.

[0235] In one possible implementation, predicting the driving trajectory of the current vehicle based on the current vehicle's state information includes: determining at least one future position of the current vehicle based on the current vehicle's state information; and predicting the driving trajectory of the current vehicle based on the at least one future position.

[0236] In one possible implementation, predicting the driving trajectory of the current vehicle based on the current vehicle's state information includes: obtaining at least one historical location of the current vehicle; determining at least one future location of the current vehicle based on the current vehicle's state information; and predicting the driving trajectory of the current vehicle based on the at least one historical location and the at least one future location.

[0237] In one possible implementation, obtaining the planned trajectory of the current vehicle and the planned trajectory of the at least one target vehicle based on the current vehicle's status information and reference line, and the status information and reference line of the at least one target vehicle, includes: for the current vehicle, determining at least one planned position of the current vehicle based on the current vehicle's status information and reference line, the at least one target vehicle's status information and reference line, and preset position intervals and speed intervals; and determining the planned trajectory of the current vehicle based on the at least one planned position.

[0238] An embodiment of this application provides a vehicle collision probability detection device, including: a processor and a memory for storing processor-executable instructions; wherein the processor is configured to implement the above-described method when executing the instructions.

[0239] Embodiments of this application provide a non-volatile computer-readable storage medium storing computer program instructions thereon, which, when executed by a processor, implement the above-described method.

[0240] Embodiments of this application provide a computer program product including computer-readable code, or a non-volatile computer-readable storage medium carrying computer-readable code, wherein when the computer-readable code is run in a processor of an electronic device, the processor in the electronic device performs the above-described method.

[0241] Embodiments of this application provide a vehicle including the aforementioned vehicle collision probability detection device. The vehicle can be a transport vehicle, such as a passenger car, bus, truck, or tractor-trailer; it can also be a special-purpose vehicle, such as a racing car, environmental protection vehicle, ambulance, or fire truck. This application does not limit the specific type of vehicle.

[0242] The vehicle collision probability detection method and apparatus described in the embodiments of this application can be applied to the fields of autonomous driving, intelligent driving, or driverless driving to improve the accuracy of vehicle collision probability detection.

[0243] Computer-readable storage media can be tangible devices capable of holding and storing instructions for use by an instruction execution device. Computer-readable storage media can be, for example—but not limited to—electrical storage devices, magnetic storage devices, optical storage devices, electromagnetic storage devices, semiconductor storage devices, or any suitable combination of the foregoing. More specific examples (a non-exhaustive list) of computer-readable storage media include: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), electrically programmable read-only memory (EPROM or flash memory), static random access memory (SRAM), compact disc read-only memory (CD-ROM), digital video disc (DVD), memory sticks, floppy disks, mechanical encoding devices, such as punch cards or recessed protrusions storing instructions thereon, and any suitable combination of the foregoing.

[0244] The computer-readable program instructions or code described herein can be downloaded from computer-readable storage media to various computing / processing devices, or downloaded via a network, such as the Internet, local area network, wide area network, and / or wireless network, to an external computer or external storage device. The network may include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. A network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards them to the computer-readable storage media in the respective computing / processing device.

[0245] The computer program instructions used to perform the operations of this application may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, status setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk, C++, etc., and conventional procedural programming languages ​​such as "e.g." or similar programming languages. The computer-readable program instructions may be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer may be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN), or it may be connected to an external computer (e.g., via the Internet using an Internet service provider). In some embodiments, electronic circuits, such as programmable logic circuits, field-programmable gate arrays (FPGAs), or programmable logic arrays (PLAs), are personalized by utilizing state information from computer-readable program instructions. These electronic circuits can execute computer-readable program instructions to implement various aspects of this application.

[0246] Various aspects of this application are described herein with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer-readable program instructions.

[0247] These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that, when executed by the processor of the computer or other programmable data processing apparatus, they create means for implementing the functions / actions specified in one or more blocks of the flowchart and / or block diagram. These computer-readable program instructions can also be stored in a computer-readable storage medium that causes a computer, programmable data processing apparatus, and / or other device to operate in a particular manner; thus, the computer-readable medium storing the instructions comprises an article of manufacture that includes instructions for implementing aspects of the functions / actions specified in one or more blocks of the flowchart and / or block diagram.

[0248] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions executed on the computer, other programmable data processing apparatus, or other device to perform the functions / actions specified in one or more boxes of a flowchart and / or block diagram.

[0249] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of apparatus, systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of an instruction containing one or more executable instructions for implementing a specified logical function. In some alternative implementations, the functions marked in the blocks may occur in a different order than those shown in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved.

[0250] It should also be noted that each block in the block diagram and / or flowchart, as well as combinations of blocks in the block diagram and / or flowchart, can be implemented using hardware (such as circuits or ASICs (Application Specific Integrated Circuits)) that performs the corresponding function or action, or using a combination of hardware and software, such as firmware.

[0251] Although the invention has been described herein in conjunction with various embodiments, those skilled in the art will understand and implement other variations of the disclosed embodiments by reviewing the accompanying drawings, disclosure, and appended claims in carrying out the claimed invention. In the claims, the word "comprising" does not exclude other components or steps, and "a" or "an" does not exclude a plurality. A single processor or other unit can implement several functions listed in the claims. While different dependent claims may recite certain measures, this does not mean that these measures cannot be combined to produce good results.

[0252] The various embodiments of this application have been described above. These descriptions are exemplary and not exhaustive, nor are they limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is chosen to best explain the principles, practical application, or improvement of the technology in the market, or to enable others skilled in the art to understand the embodiments disclosed herein.

Claims

1. A method for detecting vehicle collision probability, characterized in that, The method includes: The system acquires the current vehicle's status information and reference line, and the status information and reference line of at least one target vehicle. The status information includes at least one of lateral position, longitudinal position, speed, acceleration, heading angle, and heading angular velocity. The current vehicle's reference line includes the center line of the lane in which the current vehicle is located. The at least one target vehicle is a vehicle surrounding the current vehicle. Based on the current vehicle status information, predict the current vehicle's driving trajectory; Based on the status information of the at least one target vehicle, predict the driving trajectory of the at least one target vehicle; Based on the current vehicle's status information and reference line, and the status information and reference line of the at least one target vehicle, the planned trajectory of the current vehicle and the planned trajectory of the at least one target vehicle are obtained. The fusion of the predicted driving trajectory and the planned trajectory indicates the collision probability between the current vehicle and each target vehicle, as well as the collision probability between each target vehicle.

2. The method according to claim 1, characterized in that, The method further includes: The predicted driving trajectory of the current vehicle is fused with the planned trajectory to obtain the fused trajectory of the current vehicle. The predicted driving trajectory of the at least one target vehicle is fused with the planned trajectory to obtain the fused trajectory of the at least one target vehicle. Based on the fusion trajectory of the current vehicle and the fusion trajectory of the at least one target vehicle, the collision probability between the current vehicle and each target vehicle, and the collision probability between each target vehicle are determined.

3. The method according to claim 1 or 2, characterized in that, The method further includes: Output prompt information, which includes at least one of voice prompt information and visual prompt information, and the prompt information is used to indicate the collision probability.

4. The method according to claim 1 or 2, characterized in that, The method further includes: Based on the collision probability, corresponding driving control is performed on the current vehicle.

5. The method according to claim 2, characterized in that, The step of determining the collision probability between the current vehicle and each target vehicle, and the collision probability between each target vehicle, based on the fused trajectory of the current vehicle and the fused trajectory of at least one target vehicle, includes: A trajectory set of a first vehicle is determined, wherein the first vehicle is any vehicle among the current vehicle and the at least one target vehicle, and the trajectory set includes at least one trajectory combination, each trajectory combination including the fused trajectory of the first vehicle and the fused trajectory of a second vehicle, wherein the second vehicle is any vehicle other than the first vehicle among the current vehicle and the at least one target vehicle. Determine the cost of each trajectory combination in the trajectory set; The collision probability between the first vehicle and the second vehicle is determined based on the cost value of each trajectory combination in the trajectory set.

6. The method according to claim 5, characterized in that, Determining the cost of each trajectory combination in the trajectory set includes: Determine whether any combination of trajectories in the trajectory set collides; The combination of trajectories that do not collide is defined as the set of non-collision trajectories; Determine the cost of each trajectory combination in the set of non-collision trajectories; Determining the collision probability between the first vehicle and the second vehicle based on the cost value of each trajectory combination in the trajectory set includes: The collision probability between the first vehicle and the second vehicle is determined based on the cost value of each trajectory combination in the set of non-collision trajectories.

7. The method according to claim 6, characterized in that, The step of determining whether a collision occurs between various trajectory combinations in the trajectory set includes: For any trajectory combination, according to a preset first time series, the fused trajectory of the first vehicle and the fused trajectory of the second vehicle in the trajectory combination are discretized to obtain multiple target positions of the first vehicle and multiple target positions of the second vehicle. The first time series includes multiple time points. At any point in the first time series, it is determined whether the target area where the first vehicle is located overlaps with the target area where the second vehicle is located. The target area where the first vehicle is located corresponds to the target position of the first vehicle at that time, and the target area where the second vehicle is located corresponds to the target position of the second vehicle at that time. The overlap of target areas indicates that the trajectory combination has collided.

8. The method according to claim 7, characterized in that, The target area is a rectangular area or a three-circle area, wherein the three-circle area includes three circular areas defined by the center of the front axle, the center of the vehicle, and the center of the rear axle, respectively, and the wheelbase of the vehicle as the diameter.

9. The method according to claim 2, characterized in that, The step of fusing the predicted driving trajectory of the current vehicle with the planned trajectory to obtain the fused trajectory of the current vehicle includes: Based on the preset second time series, the predicted driving trajectory and the planned trajectory of the current vehicle are discretized to obtain multiple predicted trajectory points and multiple planned trajectory points. The fused trajectory of the current vehicle is determined based on the preset weight sequence, the multiple predicted trajectory points, and the multiple planned trajectory points.

10. The method according to claim 2, characterized in that, The step of fusing the predicted driving trajectory of the current vehicle with the planned trajectory to obtain the fused trajectory of the current vehicle includes: Determine the first polynomial coefficients corresponding to the predicted driving trajectory of the current vehicle and the second polynomial coefficients corresponding to the planned trajectory of the current vehicle. The fusion trajectory of the current vehicle is determined based on the preset coefficient weights, the first polynomial coefficients, and the second polynomial coefficients.

11. The method according to any one of claims 1, 2, 5-10, characterized in that, The step of predicting the current vehicle's trajectory based on the current vehicle's status information includes: Based on the current vehicle status information, determine at least one future location of the current vehicle; Based on the at least one future location, predict the current vehicle's trajectory.

12. The method according to any one of claims 1, 2, 5-10, characterized in that, The step of predicting the current vehicle's trajectory based on the current vehicle's status information includes: Obtain at least one historical location of the current vehicle; Based on the current vehicle status information, determine at least one future location of the current vehicle; The current vehicle's trajectory is predicted based on the at least one historical location and the at least one future location.

13. The method according to any one of claims 1, 2, 5-10, characterized in that, The step of obtaining the planned trajectory of the current vehicle and the planned trajectory of the at least one target vehicle based on the current vehicle's status information and reference line, and the status information and reference line of the at least one target vehicle, includes: For the current vehicle, at least one planned position of the current vehicle is determined based on the status information and reference line of the current vehicle, the status information and reference line of the at least one target vehicle, and the preset position interval and speed interval. The planned trajectory of the current vehicle is determined based on the at least one planned location.

14. A vehicle collision probability detection device, characterized in that, The device includes: The acquisition unit is used to acquire the status information and reference line of the current vehicle, and the status information and reference line of at least one target vehicle. The status information includes at least one of lateral position, longitudinal position, speed, acceleration, heading angle and heading angular velocity. The reference line of the current vehicle includes the center line of the lane where the current vehicle is located. The at least one target vehicle is a vehicle surrounding the current vehicle. The processing unit is configured as follows: Based on the current vehicle status information, predict the current vehicle's driving trajectory; Based on the status information of the at least one target vehicle, predict the driving trajectory of the at least one target vehicle; Based on the current vehicle's status information and reference line, and the status information and reference line of the at least one target vehicle, the planned trajectory of the current vehicle and the planned trajectory of the at least one target vehicle are obtained. The fusion of the predicted driving trajectory and the planned trajectory indicates the collision probability between the current vehicle and each target vehicle, as well as the collision probability between each target vehicle.

15. The apparatus according to claim 14, characterized in that, The processing unit is further configured to: The predicted driving trajectory of the current vehicle is fused with the planned trajectory to obtain the fused trajectory of the current vehicle. The predicted driving trajectory of the at least one target vehicle is fused with the planned trajectory to obtain the fused trajectory of the at least one target vehicle. Based on the fusion trajectory of the current vehicle and the fusion trajectory of the at least one target vehicle, the collision probability between the current vehicle and each target vehicle, and the collision probability between each target vehicle are determined.

16. The apparatus according to claim 14 or 15, characterized in that, The device further includes: An output unit is used to output prompt information, which includes at least one of voice prompt information and visual prompt information, and the prompt information is used to indicate the collision probability.

17. The apparatus according to claim 14 or 15, characterized in that, The device further includes: The control unit is used to perform corresponding driving control on the current vehicle based on the collision probability.

18. The apparatus according to claim 15, characterized in that, The step of determining the collision probability between the current vehicle and each target vehicle, and the collision probability between each target vehicle, based on the fused trajectory of the current vehicle and the fused trajectory of at least one target vehicle, includes: A trajectory set of a first vehicle is determined, wherein the first vehicle is any vehicle among the current vehicle and the at least one target vehicle, and the trajectory set includes at least one trajectory combination, each trajectory combination including the fused trajectory of the first vehicle and the fused trajectory of a second vehicle, wherein the second vehicle is any vehicle other than the first vehicle among the current vehicle and the at least one target vehicle. Determine the cost of each trajectory combination in the trajectory set; The collision probability between the first vehicle and the second vehicle is determined based on the cost value of each trajectory combination in the trajectory set.

19. The apparatus according to claim 18, characterized in that, Determining the cost of each trajectory combination in the trajectory set includes: Determine whether any combination of trajectories in the trajectory set collides; The combination of trajectories that do not collide is defined as the set of non-collision trajectories; Determine the cost of each trajectory combination in the set of non-collision trajectories; Determining the collision probability between the first vehicle and the second vehicle based on the cost value of each trajectory combination in the trajectory set includes: The collision probability between the first vehicle and the second vehicle is determined based on the cost value of each trajectory combination in the set of non-collision trajectories.

20. The apparatus according to claim 19, characterized in that, The step of determining whether a collision occurs between various trajectory combinations in the trajectory set includes: For any trajectory combination, according to a preset first time series, the fused trajectory of the first vehicle and the fused trajectory of the second vehicle in the trajectory combination are discretized to obtain multiple target positions of the first vehicle and multiple target positions of the second vehicle. The first time series includes multiple time points. At any point in the first time series, it is determined whether the target area where the first vehicle is located overlaps with the target area where the second vehicle is located. The target area where the first vehicle is located corresponds to the target position of the first vehicle at that time, and the target area where the second vehicle is located corresponds to the target position of the second vehicle at that time. The overlap of target areas indicates that the trajectory combination has collided.

21. The apparatus according to claim 20, characterized in that, The target area is a rectangular area or a three-circle area, wherein the three-circle area includes three circular areas defined by the center of the front axle, the center of the vehicle, and the center of the rear axle, respectively, and the wheelbase of the vehicle as the diameter.

22. The apparatus according to claim 15, characterized in that, The step of fusing the predicted driving trajectory of the current vehicle with the planned trajectory to obtain the fused trajectory of the current vehicle includes: Based on the preset second time series, the predicted driving trajectory and the planned trajectory of the current vehicle are discretized to obtain multiple predicted trajectory points and multiple planned trajectory points. The fused trajectory of the current vehicle is determined based on the preset weight sequence, the multiple predicted trajectory points, and the multiple planned trajectory points.

23. The apparatus according to claim 15, characterized in that, The step of fusing the predicted driving trajectory of the current vehicle with the planned trajectory to obtain the fused trajectory of the current vehicle includes: Determine the first polynomial coefficients corresponding to the predicted driving trajectory of the current vehicle and the second polynomial coefficients corresponding to the planned trajectory of the current vehicle. The fusion trajectory of the current vehicle is determined based on the preset coefficient weights, the first polynomial coefficients, and the second polynomial coefficients.

24. The apparatus according to any one of claims 14, 15, 18-23, characterized in that, The step of predicting the current vehicle's trajectory based on the current vehicle's status information includes: Based on the current vehicle status information, determine at least one future location of the current vehicle; Based on the at least one future location, predict the current vehicle's trajectory.

25. The apparatus according to any one of claims 14, 15, 18-23, characterized in that, The step of predicting the current vehicle's trajectory based on the current vehicle's status information includes: Obtain at least one historical location of the current vehicle; Based on the current vehicle status information, determine at least one future location of the current vehicle; The current vehicle's trajectory is predicted based on the at least one historical location and the at least one future location.

26. The apparatus according to any one of claims 14, 15, 18-23, characterized in that, The step of obtaining the planned trajectory of the current vehicle and the planned trajectory of the at least one target vehicle based on the current vehicle's status information and reference line, and the status information and reference line of the at least one target vehicle, includes: For the current vehicle, at least one planned position of the current vehicle is determined based on the status information and reference line of the current vehicle, the status information and reference line of the at least one target vehicle, and the preset position interval and speed interval. The planned trajectory of the current vehicle is determined based on the at least one planned location.

27. A vehicle collision probability detection device, characterized in that, include: processor; Memory used to store processor-executable instructions; The processor is configured to implement the method of any one of claims 1-13 when executing the instructions.

28. A non-volatile computer-readable storage medium storing computer program instructions thereon, characterized in that, When the computer program instructions are executed by the processor, they implement the method described in any one of claims 1-13.

29. A vehicle, characterized in that, The vehicle includes the vehicle collision probability detection device according to any one of claims 14-26, or the vehicle collision probability detection device according to claim 27.

Images