Vehicle control method and device, vehicle and storage medium

Abstract

The invention discloses a vehicle control method and device, a vehicle and a storage medium. The method comprises the following steps: determining each predicted motion track corresponding to a front vehicle and a motion probability corresponding to each predicted motion track based on a speed difference value, a motion state and lane information corresponding to the front vehicle; based on each predicted motion track corresponding to the preceding vehicle and the motion probability corresponding to each predicted motion track, determining a corresponding optimal speed sequence under the condition that the vehicle does not collide with the preceding vehicle; and controlling the vehicle to run according to the estimated discrete speed value according to the optimal speed sequence. By determining each predicted motion track corresponding to the front vehicle and the motion probability corresponding to each predicted motion track, the optimal speed sequence of the vehicle within the preset duration can be determined by referring to the influence of the motion probability corresponding to each predicted motion track of the front vehicle on the vehicle control of the vehicle; the influence of other vehicles in the environment is fully considered during vehicle control, and the vehicle control effect of the vehicle is optimized.

Description

Technical Field

[0001] The present application relates to the field of vehicle autonomous driving technology, and more specifically, to a vehicle control method, device, vehicle, and storage medium. Background Art

[0002] In the autonomous vehicle control process, if the vehicle only responds to the current speed of the preceding vehicle without predicting its intended behavior, the response will often be delayed, resulting in uncomfortable vehicle control and a significantly reduced driving experience. Therefore, it is necessary to predict the preceding vehicle's behavior and optimize the vehicle's control based on this predicted behavior to respond in advance, effectively avoiding braking or slow response. However, in general, the prediction of the preceding vehicle's behavior is limited and inaccurate, making it difficult to improve the comfort of the vehicle's control. Summary of the Invention

[0003] In view of the above problems, the present application proposes a vehicle control method, device, vehicle and storage medium, which can ensure that the influence of other vehicles in the environment is fully considered when the vehicle is controlled, and optimize the vehicle control effect of the vehicle.

[0004] In a first aspect, an embodiment of the present application provides a vehicle control method, the method comprising: determining each predicted motion trajectory corresponding to the leading vehicle and the motion probability corresponding to each predicted motion trajectory based on the speed difference, motion state, and lane information corresponding to the leading vehicle, the speed difference being the difference between the speed of the leading vehicle and the speed of a target vehicle in front of the leading vehicle, the motion probability being the probability that the leading vehicle actually travels according to the predicted motion trajectory, and the lane information comprising the steering angle between the center axis of the leading vehicle and the lane line; based on each predicted motion trajectory corresponding to the leading vehicle and the motion probability corresponding to each predicted motion trajectory, determining the optimal speed sequence corresponding to the vehicle when no collision occurs with the leading vehicle, the optimal speed sequence comprising the estimated discrete speed values ​​of the vehicle within a preset time length; and controlling the vehicle to travel according to the estimated discrete speed values ​​according to the optimal speed sequence.

[0005] In a second aspect, an embodiment of the present application provides a vehicle control device, the device comprising: a trajectory prediction module, a speed planning module, and a vehicle control module. The trajectory prediction module is used to determine each predicted motion trajectory corresponding to the preceding vehicle, and the motion probability corresponding to each predicted motion trajectory, based on the speed difference, motion state, and lane information corresponding to the preceding vehicle, wherein the speed difference is the difference between the speed of the preceding vehicle and the speed of the target vehicle in front of the preceding vehicle, and the motion probability is the probability that the preceding vehicle actually travels according to the predicted motion trajectory, and the lane information includes the steering angle between the center axis of the preceding vehicle and the lane line; the speed planning module is used to determine the optimal speed sequence corresponding to the vehicle without colliding with the preceding vehicle based on each predicted motion trajectory corresponding to the preceding vehicle, and the motion probability corresponding to each predicted motion trajectory; the optimal speed sequence includes the estimated discrete speed values ​​of the vehicle within a preset time length; and the vehicle control module is used to control the vehicle to travel according to the estimated discrete speed values ​​according to the optimal speed sequence.

[0006] In a third aspect, an embodiment of the present application provides a server comprising: one or more processors; a memory; and one or more applications, wherein the one or more applications are stored in the memory and configured to be executed by the one or more processors, and the one or more programs are configured to execute the vehicle control method provided in the first aspect above.

[0007] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, in which a program code is stored. The program code can be called by a processor to execute the vehicle control method provided in the first aspect above.

[0008] The solution provided by this application determines each predicted motion trajectory corresponding to the preceding vehicle and the motion probability corresponding to each predicted motion trajectory based on the speed difference, motion state, and lane information corresponding to the preceding vehicle. The speed difference is the difference between the speed of the preceding vehicle and the speed of a target vehicle ahead of the preceding vehicle. The motion probability is the probability that the preceding vehicle actually travels along the predicted motion trajectory. The lane information includes the steering angle between the centerline of the preceding vehicle and the lane line. Based on each predicted motion trajectory corresponding to the preceding vehicle and the motion probability corresponding to each predicted motion trajectory, the optimal speed sequence corresponding to the host vehicle without colliding with the preceding vehicle is determined. The optimal speed sequence includes estimated discrete speed values ​​of the host vehicle within a preset time period. Based on the optimal speed sequence, the host vehicle is controlled to travel according to the estimated discrete speed values. By determining each predicted motion trajectory corresponding to the preceding vehicle and the motion probability corresponding to each predicted motion trajectory, the influence of the motion probability corresponding to each predicted motion trajectory of the preceding vehicle on the control of the host vehicle can be referenced, and the optimal speed sequence for the host vehicle within the preset time period can be determined, ensuring that the influence of other vehicles in the environment is fully considered during vehicle control, thereby optimizing the vehicle control effect of the host vehicle. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0010] Figure 1 A flow chart of a vehicle control method provided in one embodiment of the present application is shown.

[0011] Figure 2 A flow chart of a vehicle control method provided in another embodiment of the present application is shown.

[0012] Figure 3 A flow chart of step S206 in another embodiment of the present application is shown.

[0013] Figure 4 A schematic diagram of an ST diagram in an embodiment of the present application is shown.

[0014] Figure 5 Another flow chart of step S206 in another embodiment of the present application is shown.

[0015] Figure 6 A schematic diagram of a process for determining the optimal speed sequence of the vehicle in an embodiment of the present application is shown.

[0016] Figure 7 A schematic structural diagram of a vehicle control device provided in an embodiment of the present application is shown.

[0017] Figure 8 A structural block diagram of a vehicle provided in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0018] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application.

[0019] During autonomous vehicle control, if the vehicle's speed planning and other controls are based solely on the current state of other vehicles in the environment without considering their intended behavior, this often results in delayed control responses, leading to uncomfortable control and a negative impact on the occupants' driving experience. Conversely, if the intended behavior of other vehicles can be accurately predicted, the vehicle can be controlled to brake or decelerate in advance, significantly improving vehicle control effectiveness.

[0020] Therefore, autonomous vehicle control typically involves predicting the behavioral intentions of other vehicles in the current environment, primarily predicting whether the vehicle ahead in the adjacent lane will turn—that is, predicting the trajectory of the preceding vehicle. However, due to the complex road conditions in real-world driving environments, it is difficult for the vehicle to fully consider the factors that influence the preceding vehicle's trajectory, resulting in a less accurate prediction of the preceding vehicle's trajectory. Without a clear trajectory, the preceding vehicle's influence on the vehicle's control becomes uncertain, significantly impacting both the accuracy of the vehicle's control and the comfort of the ride.

[0021] Therefore, the present application provides a vehicle control method, apparatus, vehicle, and storage medium. By determining each predicted motion trajectory corresponding to a preceding vehicle and the motion probability corresponding to each predicted motion trajectory, the impact of the motion probability corresponding to each predicted motion trajectory of the preceding vehicle on the vehicle control of the present vehicle can be referenced, and the optimal speed sequence of the present vehicle within a preset duration can be determined. This ensures that the influence of other vehicles in the environment is fully considered during vehicle control, thereby optimizing the vehicle control effect of the present vehicle. The specific vehicle control method is described in detail in the subsequent embodiments.

[0022] See also Figure 1 , Figure 1 A flow chart of a vehicle control method according to an embodiment of the present application is shown. Figure 1 The process shown is described in detail, and the vehicle control method may specifically include the following steps:

[0023] Step S110: Based on the speed difference, motion state, and lane information corresponding to the preceding vehicle, each predicted motion trajectory corresponding to the preceding vehicle and the motion probability corresponding to each predicted motion trajectory are determined.

[0024] In an embodiment of the present application, the speed difference is the difference between the speed of the leading vehicle and the speed of the target vehicle in front of the leading vehicle, the motion probability is the probability that the leading vehicle actually travels according to the predicted motion trajectory, and the lane information includes the steering angle between the center axis of the leading vehicle and the lane line.

[0025] The vehicle can use hardware sensing devices such as onboard cameras and lidar to obtain lane traffic flow information, such as the speed of each vehicle in the current environment and the relative distance between each vehicle and the vehicle itself. Based on this information, the vehicle in front of it can determine which other vehicles in the current environment may affect its speed planning and other vehicle control. Furthermore, based on the speed difference, motion state, and lane information corresponding to each leading vehicle, the vehicle can determine each predicted motion trajectory for each leading vehicle and the motion probability corresponding to each predicted motion trajectory. Lane information refers to the steering angle between the centerline of the leading vehicle and the lane line. Obviously, if the vehicle is traveling within the lane and does not need to turn, the steering angle between the centerline of the vehicle and the lane line should be small. However, if the vehicle needs to turn, the centerline of the vehicle will inevitably form a large angle with the lane line. Therefore, the vehicle can use the lane information corresponding to the leading vehicle to determine whether the leading vehicle intends to turn.

[0026] Specifically, the preceding vehicle in the current environment may refer to a vehicle that is in front of the vehicle, traveling in the adjacent lane, and whose relative distance to the vehicle is less than a preset distance. Obviously, if such a vehicle has the intention to change lanes, then the vehicle is more likely to have an impact on the speed planning of the vehicle during subsequent driving. Therefore, these vehicles can be regarded as the preceding vehicles corresponding to the vehicle. At the same time, based on the speed difference, motion state, and lane information corresponding to each preceding vehicle detected by the hardware sensing device of the vehicle, each predicted motion trajectory corresponding to each preceding vehicle and the motion probability corresponding to each predicted motion trajectory are determined. Obviously, the predicted motion trajectory of the preceding vehicle can characterize whether the preceding vehicle has the intention to change lanes, and the motion probability corresponding to the predicted motion trajectory can characterize the probability of the preceding vehicle changing lanes.

[0027] It is understandable that if other vehicles in the current environment are behind the vehicle, or the relative distance between them and the vehicle is greater than or equal to the preset distance, then even if the vehicle suddenly changes speed or lanes, it will not affect the speed planning of the vehicle in a short period of time. Therefore, these vehicles can be ignored as the corresponding preceding vehicles of the vehicle, that is, the impact of these vehicles on the vehicle control of the vehicle is not considered. However, it should be understood that although these vehicles may not affect the vehicle control of the vehicle, they may affect the motion trajectory of the preceding vehicle. Therefore, in order to make the predicted motion trajectory of the preceding vehicle and the corresponding motion probability determined more accurately by the vehicle, the vehicle can also obtain the speed corresponding to the target vehicle in the adjacent lane that will affect the motion trajectory of the preceding vehicle and is in front of the preceding vehicle through the hardware sensing device. In this way, the vehicle can determine the speed difference corresponding to the preceding vehicle based on the speed of the target vehicle and the speed of the preceding vehicle. Obviously, if the speed of the target vehicle is less than that of the leading vehicle, and the target vehicle is still in front of the leading vehicle, the leading vehicle is in a state of being slowed down by the target vehicle, and the probability of the leading vehicle changing lanes is higher at this time; on the contrary, if the speed of the target vehicle is greater than or equal to the speed of the leading vehicle, and the target vehicle is still in front of the leading vehicle, that is, the leading vehicle is not in a state of being slowed down, then the probability of the leading vehicle changing lanes is lower at this time.

[0028] In some embodiments, the motion state corresponding to the preceding vehicle obtained by the vehicle through the hardware sensing device may include data such as the position information, motion information, and size information of the preceding vehicle. Specifically, the vehicle's position information is the coordinate data of the vehicle at the current moment in the world coordinate system, the motion information may include data such as the vehicle's current speed and acceleration, and the size information may include data such as the length, width, and height of the vehicle. Based on this data, the vehicle can predict the motion trajectory of the preceding vehicle within a certain period of time. At the same time, the vehicle can also obtain the vehicle's motion state data through the hardware sensing device, including data such as the vehicle's current speed, acceleration, and yaw angular velocity. Based on the vehicle's motion state data, the vehicle can combine the predicted motion trajectory and motion probability of the preceding vehicle to infer the vehicle's speed planning and vehicle control within a preset period of time after the current moment.

[0029] Step S120: Based on each predicted motion trajectory corresponding to the preceding vehicle and the motion probability corresponding to each predicted motion trajectory, determine the optimal speed sequence corresponding to the vehicle without colliding with the preceding vehicle.

[0030] In this embodiment of the present application, the optimal speed sequence includes the estimated discrete speed values ​​of the vehicle within a preset duration. After obtaining each predicted motion trajectory corresponding to the preceding vehicle and the motion probability corresponding to each predicted motion trajectory, the vehicle can determine the probability of the preceding vehicle changing lanes within a preset duration after the current moment based on the predicted motion trajectory and its corresponding motion probability. It can then perform motion planning for the preceding vehicle in both the case of changing lanes and the case of not changing lanes. Furthermore, based on the different motion plans for the preceding vehicle, it can determine the corresponding speed plans for the vehicle within the preset duration. In other words, it can determine the optimal speed sequence for the vehicle in the event that the vehicle does not collide with the preceding vehicle that changes lanes.

[0031] It is understood that, after determining the predicted motion trajectory of the preceding vehicle and the motion probability corresponding to that predicted motion trajectory, the vehicle can first drive along each predicted motion trajectory to determine the corresponding speed sequence for the preceding vehicle without colliding with that predicted motion trajectory, thereby obtaining a different speed sequence corresponding to each predicted motion trajectory. Subsequently, based on the motion probability corresponding to each predicted motion trajectory (i.e., the probability that the preceding vehicle will actually drive along that predicted motion trajectory), a cost value corresponding to the speed sequence corresponding to that predicted motion trajectory can be determined. This cost value represents the probability of a collision between the preceding vehicle and the preceding vehicle if the vehicle is controlled according to that speed sequence. The speed sequence with the lowest cost value is determined as the optimal speed sequence.

[0032] The vehicle's predicted trajectory for the preceding vehicle doesn't consist of the preceding vehicle's continuous trajectory over a preset duration. Instead, it determines the position points corresponding to each discrete sampling moment within that duration. Connecting these position points into a broken line yields the predicted trajectory. Similarly, the vehicle's optimal speed sequence doesn't consist of the vehicle's continuous estimated speed values ​​over a preset duration. Instead, it determines the vehicle's estimated speed values ​​corresponding to each discrete sampling moment within that duration.

[0033] Step S130: According to the optimal speed sequence, the vehicle is controlled to travel according to the estimated discrete speed values.

[0034] In the embodiment of the present application, after the vehicle obtains the optimal speed sequence through the above steps, it can control the vehicle to travel according to the estimated discrete speed values ​​included in the optimal speed sequence for a preset duration. Of course, while traveling according to the estimated discrete speed values, the vehicle can still continuously obtain real-time information about its current motion state and the motion state of the preceding vehicle, speed difference, and lane information through hardware sensing devices. Furthermore, due to the uncontrollability of the preceding vehicle, the vehicle's determined predicted motion trajectory and its corresponding motion probability only indicate a certain probability that the preceding vehicle will travel according to the predicted motion trajectory. In reality, the preceding vehicle may temporarily change its motion trajectory. This may involve a change in speed or acceleration, or a direct change in its lane change intention. Therefore, after the vehicle determines the optimal speed sequence for a preset duration based on the preceding vehicle's predicted motion trajectory and motion probability, it does not necessarily follow the estimated discrete speed values ​​included in the optimal speed sequence. It is more likely that at some point within the preset time period, real-time updates will detect that the predicted motion trajectory and motion probability of the preceding vehicle have changed. Afterwards, the vehicle can determine a new optimal speed sequence based on the updated predicted motion trajectory and motion probability, and control the vehicle to travel according to the estimated discrete speed values ​​in the latest optimal speed sequence.

[0035] The vehicle control method provided in an embodiment of the present application determines each predicted motion trajectory corresponding to the preceding vehicle and the motion probability corresponding to each predicted motion trajectory based on the speed difference, motion state, and lane information corresponding to the preceding vehicle. The speed difference is the difference between the speed of the preceding vehicle and the speed of a target vehicle ahead of the preceding vehicle. The motion probability is the probability that the preceding vehicle actually travels according to the predicted motion trajectory. The lane information includes the steering angle between the centerline of the preceding vehicle and the lane line. Based on each predicted motion trajectory corresponding to the preceding vehicle and the motion probability corresponding to each predicted motion trajectory, the optimal speed sequence corresponding to the host vehicle without colliding with the preceding vehicle is determined. The optimal speed sequence includes estimated discrete speed values ​​of the host vehicle within a preset time period. Based on the optimal speed sequence, the host vehicle is controlled to travel according to the estimated discrete speed values. By determining each predicted motion trajectory corresponding to the preceding vehicle and the motion probability corresponding to each predicted motion trajectory, the influence of the motion probability corresponding to each predicted motion trajectory of the preceding vehicle on the vehicle control of the host vehicle can be referenced, and the optimal speed sequence of the host vehicle within the preset time period can be determined, ensuring that the influence of other vehicles in the environment is fully considered during vehicle control, thereby optimizing the vehicle control effect of the host vehicle.

[0036] See also Figure 2 , Figure 2 A flow chart of a vehicle control method according to another embodiment of the present invention is shown. Figure 2 The process shown is described in detail, and the vehicle control method may specifically include the following steps:

[0037] Step S201: If it is detected that no lane line exists in the current environment of the vehicle, a driving trajectory reference line corresponding to the vehicle is generated by fitting based on the historical motion state of the vehicle.

[0038] In the embodiment of the present application, the motion probability corresponding to each predicted motion trajectory of the preceding vehicle determined by the vehicle is used to characterize the probability that the preceding vehicle actually travels according to the predicted motion trajectory. The difference between the different predicted motion trajectories corresponding to the preceding vehicle can be whether the preceding vehicle changes lanes. Therefore, the vehicle can determine the motion probability corresponding to each predicted motion trajectory corresponding to the preceding vehicle based on whether the preceding vehicle changes lanes. It is easy to understand that if the preceding vehicle needs to change lanes, the central axis of the preceding vehicle will inevitably form a large angle with the lane line. Therefore, the vehicle can detect the lane information corresponding to the preceding vehicle in real time. However, in a complex actual driving environment, there may be some road environments where lane lines do not exist. In this case, in order to detect the lane information corresponding to the preceding vehicle and improve the accuracy of the motion probability, the vehicle can fit and generate a driving trajectory reference line corresponding to the preceding vehicle based on the historical motion state of the preceding vehicle, and based on the driving trajectory reference line, determine the virtual lane line in the current environment, and then use the angle between the central axis of the preceding vehicle and the virtual lane line as the lane information corresponding to the preceding vehicle.

[0039] During driving, the vehicle's hardware sensing device continuously captures the vehicle's current motion state, including data such as position coordinates, driving speed, and acceleration. The position coordinates can be the coordinates of the center point of the vehicle's rear axle. After the vehicle passes its current position, its motion state is not discarded but instead stored in a preset area. If the vehicle does not detect lane line data in the current environment, a corresponding driving trajectory reference line can be fitted based on the vehicle's historical motion state stored in the preset area.

[0040] Specifically, the vehicle can record its odometer value and determine the positional relationship between the two coordinate systems corresponding to the position coordinates of two adjacent historical motion states. The vehicle can project its position recorded in the historical motion state onto the current coordinate system to obtain a reference line for its driving trajectory.

[0041] Step S202: Determine a virtual lane line in the current environment based on the driving trajectory reference line.

[0042] In the embodiment of the present application, the driving trajectory reference line obtained by the vehicle can only represent the trajectory line of the center point of the rear axle of the vehicle during a period of historical time. Although this trajectory line can represent the curvature of the road to a certain extent, it cannot actually directly replace the virtual lane line. Therefore, the vehicle can also correct the driving trajectory reference line through the road curvature line and use the corrected result as the virtual lane line in the current environment. Therefore, after obtaining the driving trajectory reference line, the vehicle can also determine the road curvature radius curve based on the steering wheel angle corresponding to the vehicle during the same historical period, and comprehensively determine the curve equation corresponding to the virtual lane line based on the road curvature radius curve and the driving trajectory reference line.

[0043] Specifically, the vehicle can determine the road curvature k at a certain moment by referring to the following formula:

[0044]

[0045] Among them, δ f is the front wheel turning angle of the vehicle, is the yaw angular velocity of the vehicle, v ego That is, if the current speed of the vehicle is greater than the preset speed, it is considered that the vehicle is currently in a high-speed driving state. At this time, based on the vehicle's yaw angular velocity and the vehicle speed v ego The curvature k of the road at the current moment can be obtained more accurately; if the current speed of the vehicle is less than or equal to the preset speed, it is considered that the vehicle is currently in a low-speed driving state. At this time, the front wheel turning angle δ of the vehicle is f This allows for more accurate determination of the current road curvature k. Once the curvature k is determined, the curvature radius R = 1 / k. By combining the curvature radii corresponding to the vehicle at multiple different times, a road curvature radius curve can be generated.

[0046] In some embodiments, after obtaining the road curvature radius curve and the vehicle's driving trajectory reference line, the vehicle can use an existing solver such as Ceres Solver to solve the reference line equation y=ax 3 +bx 2 +cx, and then get the curve equation of the virtual lane line in the current environment y1=ax 3 +bx 2 +cx+lane_wide / 2 and y2=ax 3 +bx 2 +cx+lane_wide / 2, where lane_wide is the preset road width.

[0047] Step S203: Determine lane information corresponding to the preceding vehicle based on a steering angle between the center axis of the preceding vehicle and the virtual lane line.

[0048] In this embodiment of the present application, the vehicle can obtain virtual lane lines through the above steps even if no lane lines exist in the current environment. The steering angle between the center axis of the preceding vehicle and the virtual lane lines can then be used as lane information corresponding to the preceding vehicle, representing the steering degree of the preceding vehicle. Of course, if the vehicle detects lane lines in the current environment, the steering angle between the center axis of the preceding vehicle and the actual lane lines can be directly obtained.

[0049] Step S204: Based on the motion state of the preceding vehicle, the lane-changing trajectory and the non-lane-changing trajectory corresponding to the preceding vehicle are determined.

[0050] In an embodiment of the present application, the predicted motion trajectory includes a lane-changing trajectory and a non-lane-changing trajectory. The corresponding motion state of the preceding vehicle may include the preceding vehicle's speed, acceleration, heading angle, and historical information about the preceding vehicle. Based on this information, the vehicle can estimate the trajectory that the preceding vehicle would follow if it remained in its current lane without colliding with a target vehicle ahead, which is known as the non-lane-changing trajectory. It can also estimate the trajectory that the preceding vehicle would follow if it switched to its current lane without colliding with the target vehicle, which is known as the lane-changing trajectory. Specifically, based on the preceding vehicle's current fixed motion state, the vehicle can only determine one fixed lane-changing trajectory and one fixed non-lane-changing trajectory. That is, based on the preceding vehicle's current motion state, the vehicle can only determine one lane-changing trajectory and one fixed non-lane-changing trajectory. At the next moment, the preceding vehicle's motion state changes, and the target vehicle's motion state also changes. The vehicle can update the estimated lane-changing trajectory and non-lane-changing trajectory of the preceding vehicle based on the updated motion states of the preceding vehicle and the target vehicle.

[0051] Step S205: Based on the speed difference and lane information corresponding to the preceding vehicle, determine the lane-turning probability corresponding to the lane-turning trajectory and the lane-non-turning probability corresponding to the lane-non-turning trajectory. The lane-turning probability is positively correlated with the speed difference, the lane-turning probability is positively correlated with the steering angle, and the lane-non-turning probability is negatively correlated with the lane-turning probability.

[0052] In an embodiment of the present application, after determining the lane-changing trajectory and non-lane-changing trajectory of the preceding vehicle, the host vehicle can further determine the lane-changing probability of the preceding vehicle actually traveling along the lane-changing trajectory, as well as the non-lane-changing probability of the preceding vehicle actually traveling along the non-lane-changing trajectory. It is understood that the speed difference corresponding to the preceding vehicle refers to the difference between the speed of the preceding vehicle and the speed of a target vehicle located in the lane in which the preceding vehicle is located and in front of the preceding vehicle. If the difference is positive, i.e., the speed of the preceding vehicle is greater than the speed of the target vehicle, i.e., the preceding vehicle is outspeeded by the target vehicle, then it is clear that in this case, there is a greater probability that the preceding vehicle will change lanes, and the lane-changing probability will increase as the speed difference increases. Conversely, if the difference is negative, i.e., the speed of the preceding vehicle is less than the speed of the target vehicle, then the preceding vehicle is not outspeeded by the target vehicle, and in this case, the probability of the preceding vehicle changing lanes is correspondingly smaller. At the same time, if the probability of the preceding vehicle changing lanes is greater, then the probability of the preceding vehicle not changing lanes is correspondingly smaller, i.e., the non-changing probability and the lane-changing probability are negatively correlated.

[0053] In some embodiments, the lane-changing probability of the preceding vehicle may also be related to the on / off status of the preceding vehicle's turn signal. If the turn signal is on, the preceding vehicle has a higher lane-changing probability. The lane-changing probability of the preceding vehicle may also be positively correlated with the preceding vehicle's corresponding lateral speed. Obviously, if the preceding vehicle needs to change lanes, it will definitely generate a certain lateral speed. The greater the lateral speed, the greater the probability of the preceding vehicle changing lanes. Furthermore, since the lane-changing probability of the preceding vehicle is also related to the steering angle between the central axis and the lane line, as well as the speed difference of the preceding vehicle, in order to associate the final determined lane-changing probability with these parameters, the vehicle may determine the lane-changing probability P of the preceding vehicle based on the following formula:

[0054] P=1-(1-WVy)*(1-WOverlap)*(1-WIndicator)*(1-WoppreV)

[0055] Where WVy is the weight corresponding to the lateral velocity of the preceding vehicle, WOverlap is the weight corresponding to the steering angle of the preceding vehicle, WIndicator is the weight corresponding to the on / off state of the preceding vehicle's turn signal, and WoppreV is the weight corresponding to the speed difference of the preceding vehicle. If the preceding vehicle's turn signal is on, the WIndicator weight can be assigned a larger fixed value, such as 0.9 or 1; otherwise, a smaller fixed value, such as 0.1 or 0, can be assigned.

[0056] In some embodiments, the sum of the lane-changing probability and the non-lane-changing probability of the preceding vehicle may be 1. After determining the lane-changing probability of the preceding vehicle using the above method, the present vehicle may directly determine the non-lane-changing probability based on the lane-changing probability.

[0057] Step S206: discretely sample each predicted motion trajectory corresponding to the preceding vehicle and map it to the distance-time ST graph to obtain the obstacle area corresponding to each predicted motion trajectory in the ST graph.

[0058] In an embodiment of the present application, after determining each predicted motion trajectory corresponding to the preceding vehicle and its corresponding motion probability, the present vehicle can map the predicted motion trajectory of the preceding vehicle to the ST diagram, and obtain the position and duration of the lane occupied by the preceding vehicle after changing lanes, which is the obstacle area corresponding to the predicted motion trajectory of the preceding vehicle. Obviously, the obstacle area indicates that the preceding vehicle will occupy the position interval within the time period corresponding to the area. At this time, the trajectory planning of the present vehicle should not allow it to pass through the obstacle area, otherwise the present vehicle will inevitably collide with the preceding vehicle. Therefore, after mapping each predicted motion trajectory corresponding to the preceding vehicle to the same ST diagram, the present application can plan a feasible driving trajectory for the present vehicle that will not collide with the preceding vehicle based on the ST diagram with the obstacle area.

[0059] In some embodiments, a velocity planning model can be pre-built based on a partially observable Markov decision process (POMDP) ​​model. By inputting each predicted trajectory and its corresponding motion probability for the preceding vehicle into the velocity planning model, the model outputs an optimal velocity sequence for the preceding vehicle. The velocity planning model can include three main components: constructing an observation space, performing forward simulation, and calculating a cost function. Constructing the observation space involves constructing a simple ST graph and mapping the predicted trajectory of the preceding vehicle onto the ST graph to obtain the obstacle region corresponding to each predicted trajectory. Forward simulation involves, starting from the vehicle's current motion state, using the ST graph with the obstacle region mapped, and calculating the vehicle's estimated motion state at the next sampling moment based on different preset forward acceleration values, resulting in multiple simulated action sequences. Calculating the cost function involves determining the cost value for each simulated action sequence based on the motion probability corresponding to each predicted trajectory corresponding to the obstacle region, according to pre-defined cost function calculation rules.

[0060] In some embodiments, as Figure 3 As shown, the vehicle can first determine the target sampling interval for discrete sampling of each predicted motion trajectory through the following steps:

[0061] Step S2061: Obtain the collision duration between the preceding vehicle and the vehicle.

[0062] In the embodiment of the present application, the collision time determined between the preceding vehicle and the vehicle is the estimated time from the moment of collision between the preceding vehicle and the vehicle to the current moment when the vehicle is estimated to continue to move steadily according to the current operating state, including parameters such as the current position, driving speed and acceleration, and when the preceding vehicle is estimated to move according to the predicted motion trajectory. According to the length of the collision time, it can be determined whether the vehicle is at risk of collision in a short period of time. Obviously, if the collision time is shorter, the situation of the vehicle is more urgent. In this case, in order to better control the avoidance of the vehicle and the preceding vehicle, the number of samples in a short period of time can be appropriately increased to improve the accuracy of the speed planning of the vehicle in a short period of time, reduce the risk of collision, and improve the driving comfort of the vehicle.

[0063] Step S2062: If the collision duration is less than the reference duration, the target sampling interval is determined to be the first sampling interval.

[0064] Step S2063: If the collision duration is greater than or equal to the reference duration, the target sampling interval is determined to be the second sampling detection, and the first sampling interval is smaller than the second sampling interval.

[0065] In the embodiment of the present application, if the collision duration is less than the reference duration, it indicates that the vehicle and the preceding vehicle are expected to collide within the reference duration. In this case, the situation is urgent and the vehicle can increase the sampling frequency within a short period of time in the ST diagram, that is, determine the target sampling interval as the smaller first sampling interval. By increasing the sampling frequency, the vehicle's speed planning can be made more accurate within a short period of time. Figure 4 As shown, it shows a schematic diagram of the ST diagram in the embodiment of the present application. Among them, if the collision duration is less than the reference duration, the vehicle can determine the target sampling interval as the first sampling interval within the first 2 seconds of the preset duration, that is, sampling twice per second. After 2 seconds within the preset duration, the target sampling interval is determined to be the second sampling interval, that is, sampling once per second.

[0066] In some embodiments, as Figure 5 As shown, after determining the target sampling interval, the vehicle can determine the obstacle area corresponding to each predicted motion trajectory in the ST map through the following steps based on the target sampling interval:

[0067] Step S2064: Based on the target sampling interval, discrete sampling is performed on each predicted motion trajectory to obtain multiple position points corresponding to each predicted motion trajectory, and the coordinate data of each position point in the Frenet coordinate system with the road centerline as the reference is determined.

[0068] In an embodiment of the present application, after determining the target sampling interval, the vehicle can discretely sample the predicted motion trajectory of the preceding vehicle within a preset time period based on the target sampling interval to obtain a plurality of position points corresponding to each predicted motion trajectory. Each position point can represent the coordinate data of the preceding vehicle at each sampling point in the Frenet coordinate system with the center line of the road as a reference, that is, the SL coordinate system. In some embodiments, in order to better determine the relationship between each position point of the preceding vehicle and the position of the present vehicle, the coordinate data of each position point obtained by discrete sampling in the SL coordinate system corresponding to the lane where the present vehicle is located can be determined. Therefore, in the same SL coordinate system, the positional relationship between the position points obtained by discrete sampling of the predicted motion trajectory of the preceding vehicle and the current position point of the present vehicle (the origin of the SL coordinate system) can be directly determined, thereby better judging the lateral distance between each position point on the predicted motion trajectory and the present vehicle.

[0069] In some embodiments, after discrete sampling of each predicted motion trajectory corresponding to the preceding vehicle, the coordinate data of each position point in the world coordinate system is directly obtained. The host vehicle can convert the coordinate data of each position point into coordinate data in the SL coordinate system corresponding to the lane in which the host vehicle is located. According to the definition of the SL coordinate system, the S axis of the coordinate system is the lane centerline, and the L axis is perpendicular to the lane centerline. It can be understood that, assuming that the host vehicle always moves forward along the lane centerline, the L value in the coordinate data corresponding to each position point on the predicted motion trajectory of the preceding vehicle can directly represent the lateral distance between the preceding vehicle and the host vehicle.

[0070] Step S2065: Based on the coordinate data corresponding to each position point, determine whether each position point is an interference point. An interference point is a position point whose lateral distance from the vehicle is less than a preset distance.

[0071] In the embodiment of the present application, in the coordinate data of each position point in the SL coordinate system, the absolute value of the L value can directly represent the lateral distance between the front vehicle and the vehicle if the front vehicle drives to the position point. Among them, if the absolute value of the L value is less than the preset distance, it can be indicated that when the front vehicle drives to the position point, the front vehicle has not only crossed the lane line and entered the lane where the vehicle is located, but also the position of the front vehicle has blocked the straight driving route of the vehicle. In other words, if the vehicle continues to drive forward along the center line of the current lane, it will inevitably collide with the front vehicle. In this case, the position point corresponding to the front vehicle can be used as an interference point, that is, Figure 4 An endpoint of the first polyline segment in .

[0072] In some embodiments, the vehicle can determine a preset distance for determining interference points based on the vehicle's width, the vehicle's width ahead, and a preset threshold. For example, the preset distance l can be set to l = ego_half_width + obs_half_width + buffer, where ego_half_width is half the vehicle's width, obs_half_width is half the vehicle's width ahead, and buffer is the preset threshold.

[0073] Step S2066: Map the interference points in the ST map to obtain the obstacle area corresponding to each motion trajectory in the ST map.

[0074] In the embodiment of the present application, the vehicle determines all interference points in each predicted motion trajectory and obtains Figure 4 After the first polyline segment in the SL coordinate data corresponding to each interference point is determined, the second polyline segment can be determined based on the S value in the SL coordinate data. The S value of each interference point in the first polyline segment is the S value of the interference point in the SL coordinate system. The S value of each interference point in the second polyline segment can be the sum of the S value corresponding to the corresponding interference point on the first polyline segment and obs_lenght and ego_lenght, where obs_lenght is the body length of the vehicle and ego_lenght is the body length of the preceding vehicle. Figure 4 The area formed by the intersection of the first broken line segment, the second broken line segment and the two vertical lines t=1 and t=3 in the ST diagram is the obstacle area corresponding to the predicted motion trajectory in the ST diagram.

[0075] Step S207: planning a simulated action sequence for the vehicle not to pass through the obstacle area in the ST diagram, where the simulated action sequence includes the estimated discrete position data of the vehicle within a preset time period.

[0076] In the embodiment of the present application, after obtaining the obstacle area in the ST map by the above method, the vehicle can intuitively see from the ST map that within a preset time, the vehicle ahead will occupy a section of the lane. In this case, the vehicle can plan a corresponding simulated action sequence for the vehicle that will not pass through the obstacle area based on the ST map that includes the obstacle area. Figure 4 The vehicle's current position is the origin of the ST map, and the obstacle area in the ST map is the area that the vehicle predicts the vehicle ahead will occupy within the next 6 seconds. In this case, the vehicle can start forward simulation from position (0, 0) and predict its motion state at the next sampling time.

[0077] Specifically, the vehicle can determine multiple estimated motion states corresponding to the vehicle at each sampling moment based on the current motion state of the vehicle at the current moment and the preset different forward acceleration values, and obtain multiple simulated action sequences, each simulated action sequence including multiple estimated action states.

[0078] For example, three different forward acceleration values ​​can be pre-set, namely a'=-1 / 0-1. Based on the vehicle's current motion state, including its current position s, driving speed v, and acceleration a, three estimated motion states corresponding to the vehicle at the first sampling moment are determined according to forward accelerations a'=-1, a'=0, and a'=1, respectively. Based on the position s in each estimated motion state, a simulated position point s value on the ST diagram can be obtained. Subsequently, based on these three estimated motion states, multiple estimated motion states corresponding to the vehicle at the second sampling moment are determined again according to forward accelerations a'=-1, a'=0, and a'=1, respectively. The simulated position point s values ​​at adjacent sampling moments are connected into a straight line, thereby obtaining multiple simulated action sequences corresponding to the vehicle.

[0079] In some embodiments, based on the current motion state and the preset forward acceleration value, some of the simulated action sequences obtained will inevitably have their connecting lines pass through an obstacle area. Before subsequently determining the cost value corresponding to each simulated action sequence, the vehicle may first eliminate these simulated action sequences that pass through the obstacle area. This is because these simulated action sequences cannot actually be executed by the vehicle because they will inevitably collide with the vehicle ahead during driving.

[0080] Step S208: Determine the cost value corresponding to each simulated action sequence based on the motion probability corresponding to each predicted motion trajectory.

[0081] In this embodiment of the present application, after obtaining multiple simulated action sequences that avoid crossing an obstacle zone, these simulated action sequences can be further screened to select the optimal simulated action sequence as a reference for the vehicle's final driving. The selection criteria for the optimal simulated action sequence can comprehensively consider various indicators, such as ride comfort, driving efficiency, and collision risk. The vehicle can determine the cost associated with each simulated action sequence in these different aspects, ultimately obtaining a comprehensive cost for each simulated action sequence.

[0082] In some embodiments, the vehicle can comprehensively determine the ride comfort based on the differences in various parameters such as speed, acceleration, jerk, speed consistency, etc. when the vehicle travels in a speed sequence corresponding to the simulated action sequence. The smaller the difference, the smaller the speed change and the smaller the cost. The driving efficiency index can also be determined based on the distance traveled by the vehicle. The greater the distance traveled, the higher the efficiency and the smaller the cost. The collision risk index can also be determined based on the distance between the simulated action sequence and the obstacle area or the length of time stayed in the obstacle area. Obviously, the closer the distance or the longer the stay, the higher the risk of collision and the higher the cost.

[0083] Specifically, the vehicle can determine the cost value corresponding to each forward simulated action sequence based on each estimated action state included in the forward simulated action sequence and the motion probability corresponding to each motion trajectory. The motion probability corresponding to the predicted motion trajectory of the preceding vehicle represents the probability that the preceding vehicle will follow the predicted motion trajectory. In the ST diagram, the predicted motion trajectory of the preceding vehicle is represented as an obstacle area, and the motion probability corresponding to the predicted motion trajectory can be represented as the confidence level of the obstacle area in the ST diagram. The higher the confidence level of the obstacle area, the greater the probability that the preceding vehicle will follow the predicted motion trajectory, and the greater the intrusion cost corresponding to the simulated action sequence of the vehicle passing through the obstacle area.

[0084] Step S209: Determine the optimal speed sequence corresponding to the vehicle based on the simulated action sequence corresponding to the minimum cost value.

[0085] In the embodiment of the present application, the cost value corresponding to each simulated action sequence can represent the probability that the vehicle will pass through the obstacle area when driving according to the speed sequence corresponding to the simulated action sequence, and can also represent the probability of colliding with the vehicle in front. Obviously, the smaller the cost value, the safer the vehicle's driving. Therefore, after determining the cost value corresponding to each simulated action sequence, the vehicle can use the simulated action sequence with the smallest cost value as a reference for the final vehicle driving, and determine the optimal action sequence corresponding to the vehicle within a preset time based on the simulated action sequence. Among them, each estimated action state in the simulated action sequence includes data such as the estimated vehicle's driving position s, speed v, acceleration a, etc. The vehicle can directly select the speed value in each estimated action state and combine them to obtain the optimal speed sequence.

[0086] Step S210: According to the optimal speed sequence, the vehicle is controlled to travel according to the estimated discrete speed values.

[0087] In the embodiment of the present application, the specific content of step S210 can refer to the description of other embodiments and will not be repeated here.

[0088] In general, see Figure 6, which shows a flow chart of determining the optimal speed sequence for a vehicle within a preset time period, provided by an embodiment of the present application. First, the vehicle can determine whether there are lane lines in the current environment. If there are lane lines, the vehicle can directly determine the lane-turning and non-lane-turning trajectories corresponding to the preceding vehicle, as well as the corresponding lane-turning and non-lane-turning probabilities, based on the preceding vehicle's lane information, motion state, and speed difference data. If there are no lane lines, the vehicle can generate virtual lane lines in the current environment based on the vehicle's historical motion state. The vehicle can then determine the lane information corresponding to the preceding vehicle based on the virtual lane lines, and generate the lane-turning and non-lane-turning trajectories corresponding to the preceding vehicle, as well as the corresponding lane-turning and non-lane-turning probabilities, based on the preceding vehicle's speed difference and motion state data. Based on this, the vehicle can input the predicted lane-turning trajectory, lane-turning probability, non-lane-turning trajectory, and non-lane-turning probability of the preceding vehicle into a pre-constructed POMDP model to obtain the optimal speed sequence for the vehicle to avoid collision with the preceding vehicle within the preset time period.

[0089] The vehicle control method provided by the embodiment of the present application is as follows: if it is detected that there is no lane line in the current environment where the vehicle is located, then based on the historical motion state corresponding to the vehicle, a driving trajectory reference line corresponding to the vehicle is fitted to determine the virtual lane line in the current environment; based on the steering angle between the central axis of the front vehicle and the virtual lane line, the lane information corresponding to the front vehicle is determined, and based on the motion state corresponding to the front vehicle, the turning trajectory and the non-turning trajectory corresponding to the front vehicle are determined; based on the speed difference and lane information corresponding to the front vehicle, the turning probability corresponding to the turning trajectory and the non-turning probability corresponding to the non-turning trajectory are determined, each predicted motion trajectory corresponding to the front vehicle is discretely sampled and mapped into a distance-time ST map to obtain the obstacle area corresponding to each predicted motion trajectory in the ST map; a simulated action sequence for the vehicle not passing through the obstacle area is planned in the ST map, and a cost value corresponding to each simulated action sequence is determined based on the motion probability corresponding to each predicted motion trajectory; based on the simulated action sequence corresponding to the minimum cost value, the optimal speed sequence corresponding to the vehicle is determined, and the vehicle is controlled to travel according to the estimated discrete speed value. Therefore, by determining each predicted motion trajectory corresponding to the preceding vehicle and the motion probability corresponding to each predicted motion trajectory, we can refer to the impact of the motion probability corresponding to each predicted motion trajectory of the preceding vehicle on the vehicle control of this vehicle, determine the optimal speed sequence of this vehicle within a preset time length, ensure that the impact of other vehicles in the environment is fully considered during vehicle control, and optimize the vehicle control effect of this vehicle.

[0090] See also Figure 7, which shows a structural block diagram of a vehicle control device 200 provided in an embodiment of the present application. The vehicle control device 200 includes: a trajectory prediction module 210, a speed planning module 220, and a vehicle control module 230. The trajectory prediction module 210 is used to determine each predicted motion trajectory corresponding to the preceding vehicle and the motion probability corresponding to each predicted motion trajectory based on the speed difference, motion state, and lane information corresponding to the preceding vehicle. The speed difference is the difference between the speed of the preceding vehicle and the speed of the target vehicle in front of the preceding vehicle. The motion probability is the probability that the preceding vehicle actually travels according to the predicted motion trajectory. The lane information includes the steering angle between the center axis of the preceding vehicle and the lane line. The speed planning module 220 is used to determine the optimal speed sequence corresponding to the vehicle in the event of a collision with the preceding vehicle based on each predicted motion trajectory corresponding to the preceding vehicle and the motion probability corresponding to each predicted motion trajectory. The optimal speed sequence includes estimated discrete speed values ​​of the vehicle within a preset time period. The vehicle control module 230 is used to control the vehicle to travel according to the estimated discrete speed values ​​according to the optimal speed sequence.

[0091] As a possible implementation, the speed planning module 220 includes a trajectory mapping unit, an action planning unit, a cost determination unit, and a speed determination unit. The trajectory mapping unit is configured to discretely sample each predicted motion trajectory corresponding to the preceding vehicle and map it to a distance-time ST diagram, thereby obtaining the obstacle region corresponding to each predicted motion trajectory in the ST diagram. The action planning unit is configured to plan a simulated action sequence in the ST diagram that avoids the obstacle region for the vehicle itself. The simulated action sequence includes the estimated discrete position data of the vehicle itself within a preset duration. The cost determination unit is configured to determine the cost value corresponding to each simulated action sequence based on the motion probability corresponding to each predicted motion trajectory. The speed determination unit is configured to determine the optimal speed sequence for the vehicle itself based on the simulated action sequence corresponding to the minimum cost value.

[0092] As a possible implementation method, the trajectory mapping unit is also used to perform discrete sampling on each predicted motion trajectory based on the target sampling interval, obtain multiple position points corresponding to each predicted motion trajectory, and determine the coordinate data of each position point in the Frenet coordinate system with the road centerline as a reference; based on the coordinate data corresponding to each position point, determine whether each position point is an interference point, and the interference point is a position point whose lateral distance from the vehicle is less than a preset distance; map the interference point in the ST map to obtain the obstacle area corresponding to each motion trajectory in the ST map.

[0093] As a possible implementation, the speed planning module 220 further includes a duration determination unit, a first interval unit, and a second interval unit. The duration determination unit is configured to obtain the collision duration between the preceding vehicle and the vehicle; the first interval unit is configured to determine the sampling interval as the first sampling interval if the collision duration is less than a reference duration; and the second interval unit is configured to determine the sampling interval as the second sampling interval if the collision duration is greater than or equal to the reference duration, with the first sampling interval being less than the second sampling interval.

[0094] As a possible implementation method, the action planning unit is also used to determine multiple estimated motion states corresponding to the vehicle at each sampling moment based on the current motion state corresponding to the vehicle at the current moment and different preset forward acceleration values, and obtain multiple simulated action sequences, each simulated action sequence including multiple estimated action states; the cost determination unit is also used to determine the cost value corresponding to each forward simulated action sequence based on each estimated action state included in the forward simulated action sequence and the motion probability corresponding to each motion trajectory.

[0095] As a possible implementation, the trajectory prediction module 210 is further configured to determine a lane-turning trajectory corresponding to the preceding vehicle based on the corresponding motion state of the preceding vehicle; and to determine a lane-turning probability corresponding to the lane-turning trajectory and a lane-non-turning probability corresponding to the lane-non-turning trajectory based on the speed difference and lane information corresponding to the preceding vehicle, wherein the lane-turning probability is positively correlated with the speed difference, the lane-turning probability is positively correlated with the steering angle, and the non-turning probability is negatively correlated with the lane-turning probability.

[0096] As a possible implementation, the vehicle control device 200 includes a lane detection module, a lane generation module, and an angle determination module. The lane detection module is configured to, if it detects that no lane lines exist in the vehicle's current environment, generate a corresponding driving trajectory reference line based on the vehicle's corresponding historical motion state. The lane generation module is configured to determine a virtual lane line in the current environment based on the driving trajectory reference line. The angle determination module is configured to determine the lane information corresponding to the preceding vehicle based on the steering angle between the preceding vehicle's central axis and the virtual lane line.

[0097] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described devices and modules can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0098] In several embodiments provided in this application, the coupling between modules may be electrical, mechanical or other forms of coupling.

[0099] In addition, the functional modules in the various embodiments of the present application may be integrated into a processing module, or each module may exist physically separately, or two or more modules may be integrated into a single module. The above-mentioned integrated modules may be implemented in the form of hardware or software functional modules.

[0100] In summary, the solution provided by this application determines each predicted motion trajectory corresponding to the preceding vehicle and the motion probability corresponding to each predicted motion trajectory based on the speed difference, motion state, and lane information corresponding to the preceding vehicle. The speed difference is the difference between the speed of the preceding vehicle and the speed of the target vehicle ahead of the preceding vehicle. The motion probability is the probability that the preceding vehicle actually travels along the predicted motion trajectory. The lane information includes the steering angle between the centerline of the preceding vehicle and the lane line. Based on each predicted motion trajectory corresponding to the preceding vehicle and the motion probability corresponding to each predicted motion trajectory, the optimal speed sequence corresponding to the host vehicle without colliding with the preceding vehicle is determined. The optimal speed sequence includes the estimated discrete speed values ​​of the host vehicle within a preset time period. Based on the optimal speed sequence, the host vehicle is controlled to travel according to the estimated discrete speed values. By determining each predicted motion trajectory corresponding to the preceding vehicle and the motion probability corresponding to each predicted motion trajectory, the influence of the motion probability corresponding to each predicted motion trajectory of the preceding vehicle on the control of the host vehicle can be referenced, and the optimal speed sequence for the host vehicle within the preset time period can be determined, ensuring that the influence of other vehicles in the environment is fully considered during vehicle control, thereby optimizing the vehicle control effect of the host vehicle.

[0101] See also Figure 8 , which shows a structural block diagram of a vehicle 400 provided in an embodiment of the present application. The vehicle 400 in the present application may include one or more of the following components: a processor 410, a memory 420, and one or more application programs, wherein the one or more application programs may be stored in the memory 420 and configured to be executed by the one or more processors 410, and the one or more programs are configured to execute the method described in the aforementioned method embodiment.

[0102] Processor 410 may include one or more processing cores. Processor 410 utilizes various interfaces and circuits to connect various components within the computer device. It executes instructions, programs, code sets, or instruction sets stored in memory 420, as well as accesses data stored in memory 420, to perform various functions of the computer device and process data. Optionally, processor 410 may be implemented using at least one of the following hardware forms: digital signal processing (DSP), field-programmable gate array (FPGA), and programmable logic array (PLA). Processor 410 may integrate one or a combination of a central processing unit (CPU), a graphics processing unit (GPU), and a modem. The CPU primarily processes the operating system, user interface, and application programs; the GPU is responsible for rendering and drawing display content; and the modem handles wireless communications. It is understood that the modem may not be integrated into processor 410 and may be implemented separately via a communications chip.

[0103] The memory 420 may include a random access memory (RAM) or a read-only memory (ROM). The memory 420 may be used to store instructions, programs, codes, code sets, or instruction sets. The memory 420 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for implementing at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the following various method embodiments, etc. The data storage area may also store data created by the computer device during use (such as a phone book, audio and video data, chat history data, etc.).

[0104] The embodiment of the present application provides a structural block diagram of a computer-readable storage medium, wherein the computer-readable storage medium stores program code, which can be called by a processor to execute the method described in the above method embodiment.

[0105] The computer-readable storage medium may be an electronic memory such as a flash memory, an EEPROM (Electrically Erasable Programmable Read-Only Memory), an EPROM, a hard disk, or a ROM. Alternatively, the computer-readable storage medium includes a non-transitory computer-readable storage medium. The computer-readable storage medium has storage space for program codes for executing any of the method steps described above. These program codes can be read from or written to one or more computer program products. The program codes can be compressed, for example, in an appropriate form.

[0106] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A vehicle control method, characterized in that: The method comprises: Determining each predicted motion trajectory corresponding to the leading vehicle and a motion probability corresponding to each predicted motion trajectory based on a speed difference, motion state, and lane information corresponding to the leading vehicle, wherein the speed difference is the difference between the speed of the leading vehicle and the speed of a target vehicle ahead of the leading vehicle, the motion probability is the probability that the leading vehicle actually travels along the predicted motion trajectory, and the lane information includes a steering angle between the centerline of the leading vehicle and the lane line; Based on each predicted motion trajectory corresponding to the preceding vehicle and the motion probability corresponding to each predicted motion trajectory, determining an optimal speed sequence corresponding to the host vehicle without colliding with the preceding vehicle, the optimal speed sequence including estimated discrete speed values ​​of the host vehicle within a preset time period; According to the optimal speed sequence, the vehicle is controlled to travel according to the estimated discrete speed values.

2. The method according to claim 1, characterized in that The obtaining, based on each predicted motion trajectory corresponding to the preceding vehicle and the motion probability corresponding to each predicted motion trajectory, of an optimal speed sequence corresponding to the vehicle without colliding with the preceding vehicle includes: Performing discrete sampling on each of the predicted motion trajectories corresponding to the preceding vehicle and mapping the sampled ... Planning a simulated action sequence for the vehicle not to pass through the obstacle area in the ST diagram, the simulated action sequence including estimated discrete position data of the vehicle within a preset time period; Determining a cost value corresponding to each of the simulated action sequences based on a motion probability corresponding to each of the predicted motion trajectories; Based on the simulated action sequence corresponding to the minimum cost value, an optimal speed sequence corresponding to the host vehicle is determined.

3. The method according to claim 2, characterized in that The step of discretely sampling each predicted motion trajectory corresponding to the preceding vehicle and mapping the sampled ... Based on a target sampling interval, discretely sampling each of the predicted motion trajectories is performed to obtain a plurality of position points corresponding to each of the predicted motion trajectories, and determining coordinate data of each of the position points in a Frenet coordinate system with a road centerline as a reference; Based on the coordinate data corresponding to each position point, determining whether each position point is an interference point, wherein the interference point is a position point whose lateral distance from the host vehicle is less than a preset distance; The interference points are mapped in the ST map to obtain the obstacle area corresponding to each motion trajectory in the ST map.

4. The method according to claim 3, characterized in that Before discretely sampling each predicted motion trajectory based on a target sampling interval to obtain a plurality of position points corresponding to each predicted motion trajectory, the method further includes: Obtaining the collision duration between the preceding vehicle and the own vehicle; If the collision duration is less than the reference duration, determining the target sampling interval as the first sampling interval; If the collision duration is greater than or equal to the reference duration, the target sampling interval is determined to be the second sampling detection, and the first sampling interval is smaller than the second sampling interval.

5. The method according to claim 2, characterized in that The step of planning a simulated action sequence in which the vehicle does not pass through the obstacle area in the ST diagram includes: Based on the current motion state of the host vehicle at the current moment and different preset forward acceleration values, multiple estimated motion states corresponding to the host vehicle at each sampling moment are determined to obtain multiple simulated action sequences, each of the simulated action sequences including multiple estimated action states; The determining, based on the motion probability corresponding to each predicted motion trajectory, a cost value corresponding to each simulated action sequence includes: Based on each of the estimated action states included in the forward simulation action sequence and the motion probability corresponding to each of the motion trajectories, a cost value corresponding to each of the forward simulation action sequences is determined.

6. The method according to any one of claims 1 to 5, characterized in that The predicted motion trajectory includes a lane-changing trajectory and a non-lane-changing trajectory; determining each predicted motion trajectory corresponding to the preceding vehicle and a motion probability corresponding to each predicted motion trajectory based on a speed difference, a motion state, and lane information corresponding to the preceding vehicle includes: Determining a lane-changing trajectory and a non-lane-changing trajectory corresponding to the preceding vehicle based on a motion state corresponding to the preceding vehicle; Based on the speed difference and lane information corresponding to the leading vehicle, a lane-turning probability corresponding to the lane-turning trajectory and a lane-non-turning probability corresponding to the lane-non-turning trajectory are determined, wherein the lane-turning probability is positively correlated with the speed difference, the lane-turning probability is positively correlated with the steering angle, and the lane-non-turning probability is negatively correlated with the lane-turning probability.

7. The method according to any one of claims 1 to 5, characterized in that Before determining each predicted motion trajectory corresponding to the preceding vehicle and the motion probability corresponding to each predicted motion trajectory based on the speed difference, motion state, and lane information corresponding to the preceding vehicle, the method further includes: If it is detected that there is no lane line in the current environment of the vehicle, then based on the historical motion state of the vehicle, a driving trajectory reference line corresponding to the vehicle is generated by fitting; Determining a virtual lane line in the current environment based on the driving trajectory reference line; Lane information corresponding to the leading vehicle is determined based on a steering angle between a center axis of the leading vehicle and the virtual lane line.

8. A vehicle control device, characterized in that: The device comprises: a trajectory prediction module for determining each predicted motion trajectory corresponding to the preceding vehicle and a motion probability corresponding to each predicted motion trajectory based on a speed difference, motion state, and lane information corresponding to the preceding vehicle, wherein the speed difference is the difference between the speed of the preceding vehicle and the speed of a target vehicle ahead of the preceding vehicle, the motion probability is the probability that the preceding vehicle actually travels along the predicted motion trajectory, and the lane information includes a steering angle between the centerline of the preceding vehicle and the lane line; a speed planning module, configured to determine, based on each predicted motion trajectory corresponding to the preceding vehicle and the motion probability corresponding to each predicted motion trajectory, an optimal speed sequence corresponding to the host vehicle without colliding with the preceding vehicle, the optimal speed sequence comprising estimated discrete speed values ​​of the host vehicle within a preset time period; A vehicle control module is used to control the vehicle to travel according to the estimated discrete speed values ​​based on the optimal speed sequence.

9. A vehicle, characterized in that: The vehicle comprises: one or more processors; Memory; One or more applications, wherein the one or more applications are stored in the memory and configured to be executed by the one or more processors, and the one or more applications are configured to execute the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores program code, which can be called by a processor to execute the method according to any one of claims 1 to 7.

Images