Methods, apparatus, devices, media, and products for determining reasonableness of predicted trajectories

Abstract

The method, device, equipment, medium and product for determining the rationality of the predicted trajectory are disclosed. The first predicted trajectory corresponding to a first target in a future preset time period is obtained, and the second predicted trajectory corresponding to a second target in the future preset time period is obtained. Then, at least one first vector corresponding to at least one first trajectory point in the first predicted trajectory is determined. At least one second vector corresponding to at least one second trajectory point in the second predicted trajectory is determined. Then, based on the at least one first vector and the at least one second vector, the rationality information of at least one of the first predicted trajectory and the second predicted trajectory is determined. Thus, the rationality of the predicted trajectory is more accurately and reasonably determined, so that more accurate and reasonable trajectory prediction can be performed subsequently.

Description

Technical Field

[0001] This disclosure relates to the field of computer technology, and in particular to a method, apparatus, device, medium, and product for determining the reasonableness of a predicted trajectory. Background Technology

[0002] Trajectory prediction refers to the technique of analyzing the motion characteristics of a target, given its past or current trajectory, to deduce its possible future trajectory over a period of time. For example, in the field of autonomous driving, analyzing the motion characteristics of a vehicle can predict its possible future trajectory.

[0003] With the advancement of computer vision technology, trajectory prediction has demonstrated its importance in fields such as autonomous driving, surveillance, crowd simulation, and mobile robot navigation. Summary of the Invention

[0004] Embodiments of this disclosure provide a method, apparatus, device, medium, and product for determining the reasonableness of a predicted trajectory.

[0005] According to one aspect of the present disclosure, a method for determining the reasonableness of a predicted trajectory is provided, comprising:

[0006] Obtain the first predicted trajectory of the first target within a future preset time period, and obtain the second predicted trajectory of the second target within the same future preset time period;

[0007] Determine at least one first vector corresponding to at least one first trajectory point in the first predicted trajectory;

[0008] Determine at least one second vector corresponding to at least one second trajectory point in the second predicted trajectory;

[0009] Based on the at least one first vector and the at least one second vector, determine the plausibility information of at least one of the first predicted trajectory and the second predicted trajectory.

[0010] According to another aspect of the present disclosure, an apparatus for determining the reasonableness of a predicted trajectory is provided, comprising:

[0011] The trajectory prediction module is used to obtain a first predicted trajectory of a first target within a future preset time period, and to obtain a second predicted trajectory of a second target within the same future preset time period.

[0012] The first vector module is used to determine at least one second vector corresponding to at least one second trajectory point in the second predicted trajectory;

[0013] The second vector module is used to determine at least one second vector corresponding to at least one second trajectory point in the second predicted trajectory;

[0014] A rationality determination module is used to determine the rationality information of at least one of the first predicted trajectory and the second predicted trajectory based on the at least one first vector and the at least one second vector.

[0015] According to another aspect of the present disclosure, an electronic device is provided, comprising:

[0016] Memory, used to store computer program products;

[0017] A processor is configured to execute a computer program product stored in the memory, and when the computer program product is executed, to implement the method for determining the reasonableness of a predicted trajectory as described in any of the above embodiments.

[0018] According to another aspect of the present disclosure, a computer-readable storage medium is provided that stores computer program instructions thereon, which, when executed by a processor, implement the method for determining the reasonableness of a predicted trajectory as described in any of the above embodiments.

[0019] According to another aspect of the present disclosure, a computer program product is provided, including computer program instructions, characterized in that, when executed by a processor, the computer program instructions implement the method for determining the reasonableness of a predicted trajectory as described in any of the above embodiments.

[0020] Based on the embodiments of this disclosure, by obtaining a first predicted trajectory corresponding to a first target within a future preset time period and obtaining a second predicted trajectory corresponding to a second target within the same future preset time period, at least one first vector corresponding to at least one first trajectory point in the first predicted trajectory is determined; at least one second vector corresponding to at least one second trajectory point in the second predicted trajectory is determined; and then, based on at least one first vector and at least one second vector, the rationality information of at least one of the first and second predicted trajectories is determined. In this embodiment, the rationality of the predicted trajectories corresponding to two targets is determined by the vectors corresponding to the two targets. Since the vectors combine multiple trajectory points, and each trajectory point corresponds to a time point, the rationality of the predicted trajectory is determined by combining the time dimension. Therefore, when determining the rationality of the predicted trajectory, this embodiment considers not only whether the two targets will collide and the time of collision, but also whether there is a reasonable buffer range for avoidance between the two targets. This allows for a more accurate and reasonable determination of the rationality of the predicted trajectory, facilitating more accurate and reasonable trajectory prediction subsequently, and enabling safe vehicle driving control based on the predicted trajectory, thereby improving vehicle driving safety.

[0021] The technical solutions of this disclosure will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0022] The accompanying drawings, which form part of this specification, illustrate embodiments of this disclosure and, together with the description, serve to explain the principles of this disclosure.

[0023] This disclosure will become clearer with reference to the accompanying drawings and the following detailed description, wherein:

[0024] Figure 1 This is a flowchart illustrating a method for determining the reasonableness of a predicted trajectory provided in an exemplary embodiment of this disclosure;

[0025] Figure 2a This is a schematic diagram of the trajectory of the first target a yielding to the second target b in an optional example of this disclosure;

[0026] Figure 2b This is a schematic diagram of the trajectory of the second target b yielding to the first target a in another optional example of this disclosure;

[0027] Figure 2c This is a schematic diagram of the trajectory of the collision between the first target a and the second target b in yet another optional example of this disclosure;

[0028] Figure 3 This is a flowchart illustrating the process of determining the first vector in a method for determining the reasonableness of a predicted trajectory, provided in another exemplary embodiment of this disclosure.

[0029] Figure 4 This is a flowchart illustrating the process of determining the second vector in a method for determining the reasonableness of a predicted trajectory, provided in yet another exemplary embodiment of this disclosure.

[0030] Figure 5a This is a flowchart illustrating the process of determining reasonableness information in a method for determining the reasonableness of a predicted trajectory, provided in an exemplary embodiment of this disclosure.

[0031] Figure 5b This is a schematic diagram of the trajectory where the first vector and the second vector intersect in another optional example of this disclosure;

[0032] Figure 6 This is a schematic diagram of the structure of an apparatus for determining the reasonableness of a predicted trajectory provided in an exemplary embodiment of this disclosure;

[0033] Figure 7 This is a schematic diagram of the structure of an apparatus for determining the reasonableness of a predicted trajectory provided in another exemplary embodiment of this disclosure;

[0034] Figure 8 This is a schematic diagram of the structure of an apparatus for determining the reasonableness of a predicted trajectory, provided in yet another exemplary embodiment of this disclosure;

[0035] Figure 9 A block diagram of an electronic device according to an embodiment of the present disclosure is shown. Detailed Implementation

[0036] Hereinafter, exemplary embodiments according to the present disclosure will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of the present disclosure, and not all embodiments of the present disclosure, and it should be understood that the present disclosure is not limited to the exemplary embodiments described herein.

[0037] It should be noted that, unless otherwise specifically stated, the relative arrangement, numerical expressions, and values ​​of the components and steps set forth in these embodiments do not limit the scope of this disclosure.

[0038] Those skilled in the art will understand that the terms "first," "second," etc., in the embodiments of this disclosure are only used to distinguish different steps, devices, or modules, and do not represent any specific technical meaning, nor do they indicate a necessary logical order between them.

[0039] It should also be understood that in the embodiments disclosed herein, "a plurality of" may refer to two or more, and "at least one" may refer to one, two or more.

[0040] It should also be understood that any component, data or structure mentioned in the embodiments of this disclosure can generally be understood as one or more unless expressly defined or given to the contrary in the context.

[0041] Furthermore, the term "and / or" in this disclosure is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this disclosure generally indicates that the preceding and following related objects have an "or" relationship. The data referred to in this disclosure can include unstructured data such as text, images, and videos, as well as structured data.

[0042] It should also be understood that the description of the various embodiments in this disclosure emphasizes the differences between the various embodiments, and the similarities or similarities can be referred to each other. For the sake of brevity, they will not be described in detail.

[0043] At the same time, it should be understood that, for ease of description, the dimensions of the various parts shown in the accompanying drawings are not drawn according to actual scale.

[0044] The following description of at least one exemplary embodiment is merely illustrative and is in no way intended to limit this disclosure or its application or use.

[0045] Techniques, methods, and equipment known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and equipment should be considered part of the specification.

[0046] It should be noted that similar labels and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be discussed further in subsequent figures.

[0047] The embodiments disclosed herein can be applied to electronic devices such as terminal devices, computer systems, and servers, and can operate together with a wide range of other general-purpose or special-purpose computing system environments or configurations. Examples of well-known terminal devices, computing systems, environments, and / or configurations suitable for use with electronic devices such as terminal devices, computer systems, and servers include, but are not limited to: personal computer systems, server computer systems, thin clients, thick clients, handheld or laptop devices, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputer systems, mainframe computer systems, and distributed cloud computing environments including any of the above systems, etc.

[0048] Electronic devices such as terminal devices, computer systems, and servers can be described in the general context of computer system executable instructions (such as program modules) executed by a computer system. Typically, program modules can include routines, programs, object programs, components, logic, data structures, etc., which perform specific tasks or implement specific abstract data types. Computer systems / servers can be implemented in distributed cloud computing environments, where tasks are executed by remote processing devices linked through communication networks. In distributed cloud computing environments, program modules can reside on local or remote computing system storage media, including storage devices.

[0049] Application Overview

[0050] When trajectory prediction is applied in autonomous driving scenarios, it predicts the behavior of road traffic participants (vehicles, pedestrians, etc.) several seconds into the future. This can be represented by the locations they will pass through over a period of time (i.e., their trajectories), and the reasonableness of the trajectory is an important criterion for evaluating the quality of the predicted trajectory. To improve the effectiveness of trajectory prediction, it is necessary to optimize the trajectory prediction model during training and to evaluate the already trained trajectory prediction model.

[0051] During trajectory prediction, when there are two traffic participants with an interactive relationship (e.g., a car approaching an intersection and another car preparing to exit the intersection and merge into the lane, or a motor vehicle preparing to start at a crosswalk and a pedestrian preparing to cross the crosswalk on the roadside), a reasonable interactive trajectory should not only avoid the risk of collision, but also maintain a safe and courteous distance from other traffic participants.

[0052] For two traffic participants a and b with an interactive relationship, given their predicted trajectories traj_a (corresponding to traffic participant a) and traj_b (corresponding to traffic participant b), the time interval between any two trajectory points in the predicted trajectory is 0.1 seconds. When predicting a trajectory of 9 seconds, 91 trajectory points can be obtained, with the order of the trajectory points from i = 0 to 90. In the process of implementing this disclosure, the inventors discovered through research that in related technologies, the rationality of the interactive trajectory is often evaluated based on whether the distance between the trajectories of the two traffic participants at corresponding time points will collide. For example, in the above example, the distance between trajectory point p_a_i of traffic participant a at point i and trajectory point p_b_i of traffic participant b is calculated. For example, the coordinates of trajectory point p_a_i are... The coordinates of the trajectory point p_b_i are At that time, it can be based on The distance is calculated, and if the distance is less than a given threshold, it is considered that the two trajectories have collided. That is, the relevant technology determines the rationality of the predicted trajectory by the distance between the trajectory points, without taking into account the reasonable yielding buffer distance (for example, the yielding vehicle in the interaction does not pass right next to the vehicle in front, but leaves enough buffer distance in space). Therefore, determining the rationality of the predicted trajectory solely by the distance between the trajectory points of two traffic participants does not conform to the actual traffic situation.

[0053] Exemplary methods

[0054] Figure 1 This is a flowchart illustrating a method for determining the reasonableness of a predicted trajectory according to an exemplary embodiment of this disclosure. This embodiment can be applied to electronic devices, such as... Figure 1 As shown, it includes the following steps:

[0055] Step 102: Obtain the first predicted trajectory of the first target within a preset time period in the future, and obtain the second predicted trajectory of the second target within a preset time period in the future.

[0056] In this context, the first target and the second target are two traffic participants (e.g., vehicles, pedestrians, etc.) in the current application scenario. Optionally, the first predicted trajectory corresponding to the first target and the second predicted trajectory corresponding to the second target can be obtained based on any prediction method. The prediction methods corresponding to the first and second predicted trajectories can be the same or different. A trajectory is a set of coordinate points of an object (target) over a time series. For example, a trajectory traj = {p0, p1, ..., pi, ..., p90}, where p_i = (t_i, x_i, y_i, theta_i) is also called a trajectory point. t represents time; x and y represent the coordinates of the object in the global coordinate system at that time point; theta represents the direction of movement of the object at that time point. Since the multiple trajectory points included in a trajectory have an order and a preset time interval (e.g., 0.1 seconds), the trajectory point can be simplified as p_i = (x_i, y_i). Optionally, the first predicted trajectory includes multiple first trajectory points, and the second predicted trajectory includes multiple second trajectory points. Aligned in the time dimension, each first trajectory point can be determined to correspond to a second trajectory point. For example, the i-th first trajectory point p_a_i corresponds to the i-th second trajectory point p_b_i, where i ranges from 0 to n, and n is the number of first trajectory points included in the first predicted trajectory (or the number of second trajectory points included in the second predicted trajectory) minus one.

[0057] Step 104: Determine at least one first vector corresponding to at least one first trajectory point in the first predicted trajectory.

[0058] In some optional embodiments, if the movement vectors of two traffic participants (corresponding to the first and second targets in this embodiment) intersect within a time interval (e.g., 0.1 seconds) between two adjacent trajectory points at the same moment (the moment corresponding to any trajectory point in the predicted trajectory), then it can be deduced that the two traffic participants collided at that moment. However, analysis of a large amount of real traffic data shows that in real-world scenarios, traffic participants exhibit avoidance behaviors, thereby reducing the occurrence of collisions. For example, as... Figure 2a As shown, the first objective a yields to the second objective b, and the second objective b arrives first at the point of conflict between the trajectories (i.e., the intersection of the trajectories of the two traffic participants); for example, as... Figure 2b As shown, the second target b yields to the first target a, and the first target a reaches the conflict point between trajectories first; however, in some special cases, a collision may occur, for example, as... Figure 2c As shown, the first target a and the second target b arrive at the collision point between their trajectories at the same time, and the first target a collides with the second target b. Figures 2a-2cIn the diagram, p_a_i represents the first trajectory point corresponding to the first target at time i, p_a_i+1 represents the first trajectory point corresponding to the first target at time i+1, p_b_i represents the second trajectory point corresponding to the second target at time i, and p_b_i+1 represents the second trajectory point corresponding to the second target at time i+1.

[0059] As can be seen from the above examples, collision recognition between two traffic participants requires an extension to the time dimension to incorporate reasonable yielding time into trajectory prediction. Therefore, this disclosure proposes replacing the first trajectory point with a first vector for trajectory reasonableness assessment, thereby making the reasonableness assessment results more accurate through the extension of the time dimension.

[0060] Optionally, at least one first trajectory point included in the first predicted trajectory is used as a reference point, and at least one first trajectory point related to the reference point is determined to jointly determine the first vector corresponding to the reference point. The first vector not only reflects the position of the reference point at the corresponding time, but also reflects the reasonable avoidance space and time required to avoid collision with the first target at that time. Therefore, the first vector is used to replace the first trajectory point to determine the trajectory.

[0061] Step 106: Determine at least one second vector corresponding to at least one second trajectory point in the second predicted trajectory.

[0062] In this embodiment, the method of determining the second vector is the same as that of the first vector. The second vector is determined by the same number of second trajectory points as the first vector. The second vector represents the trajectory of the second target at a certain moment, which not only reflects the position of the second target at that moment, but also reflects the reasonable avoidance space and time required to avoid collision with the second target at that moment.

[0063] In this embodiment of the disclosure, the execution of steps 104 and 106 is not strictly ordered. Step 104 may be executed first, followed by step 106; or step 106 may be executed first, followed by step 104; or steps 104 and 106 may be executed simultaneously.

[0064] Step 108: Based on at least one first vector and at least one second vector, determine the plausibility information of at least one of the first predicted trajectory and the second predicted trajectory.

[0065] Optionally, the first trajectory points in the first predicted trajectory and the second trajectory points in the second predicted trajectory correspond one-to-one in the time dimension, and each first vector corresponds to one first trajectory point. Therefore, each first vector corresponds to one second vector. In this embodiment of the present disclosure, the reasonableness of at least one of the first predicted trajectory and the second predicted trajectory can be determined by determining whether there is an intersection between each first vector and its corresponding second vector. For example, if there is at least one intersection between a first vector and its corresponding second vector, it indicates that the first predicted trajectory and the second predicted trajectory have an intersection at a certain prediction time (i.e., a collision will occur when the first target and the second target travel along the first predicted trajectory and the second predicted trajectory, respectively). At this time, it can be determined that the first predicted trajectory and / or the second predicted trajectory are unreasonable, that is, the reasonableness information is unreasonable.

[0066] The method for determining the rationality of a predicted trajectory provided in the above embodiments of this disclosure involves obtaining a first predicted trajectory corresponding to a first target within a preset future time period, and obtaining a second predicted trajectory corresponding to a second target within the same preset future time period. Then, at least one first vector corresponding to at least one first trajectory point in the first predicted trajectory is determined; at least one second vector corresponding to at least one second trajectory point in the second predicted trajectory is determined; and finally, based on at least one first vector and at least one second vector, rationality information for at least one of the first and second predicted trajectories is determined. In this embodiment, the rationality of the predicted trajectories corresponding to two targets is determined by using vectors corresponding to two targets. Since the vectors combine multiple trajectory points, and each trajectory point corresponds to a time point, the rationality of the predicted trajectory is determined by combining the time dimension. Therefore, when determining the rationality of the predicted trajectory, this embodiment considers not only whether the two targets will collide and the time of collision, but also whether there is a reasonable buffer range for avoidance between the two targets. This allows for a more accurate and reasonable determination of the rationality of the predicted trajectory, facilitating more accurate and reasonable trajectory prediction subsequently, and enabling safe vehicle driving control based on the predicted trajectory, thereby improving vehicle driving safety.

[0067] like Figure 3 As shown above, in the above Figure 1 Based on the illustrated embodiment, step 104 may include the following steps:

[0068] Step 1041: Using the first trajectory point as the base point, determine a first number of first trajectory points in the first predicted trajectory.

[0069] In some optional examples, with the first trajectory point as the center trajectory point, a second number of first trajectory points adjacent to the center trajectory point are determined in two directions in the first preset trajectory; the two sets of the second number of first trajectory points and the center trajectory point are used as the first number of first trajectory points.

[0070] During the target's movement, forward and backward avoidance maneuvers are required. Therefore, for a given moment in the first predicted trajectory (e.g., moment i), a first number of first trajectory points are obtained through a time window of a certain length. For example, given a time window of 2k, for the first target a, taking the i-th trajectory point Ti corresponding to moment i in the first predicted trajectory as the center trajectory point, extending k points forward and backward along the first predicted trajectory (corresponding to the second number) respectively, we can obtain two edge trajectory points corresponding to the time window, Ti-k (not less than 0) and Ti+k. Connecting the two edge trajectory points Ti-k to Ti+k yields the first vector. (Different values ​​of i can yield different first vectors). For the first target, the reasonable time to give way when the first target passes in front is k trajectory points (the value of k can be determined according to the specific application scenario or according to big data statistics), that is, trajectory points i to i+k. The reasonable time to avoid the first target passing behind is k trajectory points, that is, trajectory points ik to i.

[0071] In some other optional examples, a third number of first trajectory points adjacent to the starting trajectory point are determined along a first preset trajectory, with the first trajectory point as the starting trajectory point; the third number of first trajectory points and the starting trajectory point are used as the first number of first trajectory points.

[0072] Since the first predicted trajectory is a predicted travel path for a certain period of time after the first target, the forward direction of the first predicted trajectory is the forward direction of the first target. Collisions between traffic participants typically occur in the forward direction. This example focuses on the forward direction, using the first trajectory point corresponding to a preset time (time i) in the first predicted trajectory as the starting trajectory point. A third number of first trajectory points adjacent to this starting trajectory point are determined along the forward direction of the first preset trajectory. For example, the third number is denoted as j, corresponding to a time window of j. Using the i-th trajectory point Ti corresponding to time i in the first predicted trajectory as the starting trajectory point, j first trajectory points are obtained forward along the first predicted trajectory. The edge trajectory point corresponding to the time window is Ti+j. Connecting the starting trajectory point to the edge trajectory points Ti to Ti+j yields the first vector. (Different values ​​of i can yield different first vectors), where for the first target, the reasonable yield time for the first target to pass in front is j trajectory points (the value of j can be determined according to the specific application scenario or according to big data statistics), that is, trajectory points from i to i+j.

[0073] Step 1042: Determine the first vector based on the first number of first trajectory points.

[0074] In this embodiment, after determining a first number of first trajectory points, connecting any two of these first trajectory points yields a first vector. In some optional examples, the first vector is determined by connecting the first trajectory points that are sorted first and last in the first predicted trajectory. Optionally, the process of determining the first vector may include:

[0075] The starting point and ending point of the first vector are determined by the order of the first trajectory points in the first preset trajectory, and the first vector is determined based on the starting point and ending point of the first vector.

[0076] The first predicted trajectory can be decomposed into multiple first vectors by the above method of determining the first vector. For example, when the first trajectory includes 90 first trajectory points, 90-2k or 90-j first vectors can be obtained. The number of first vectors is less than the number of first trajectory points, which reduces the computational complexity of rationality information and the amount of computation required for rationality information.

[0077] Optionally, the starting point and the ending point of the first vector are connected to obtain a first vector from the starting point to the ending point of the first vector. This first vector expresses the position of the corresponding base point in the first predicted trajectory and the avoidance space. Since the avoidance space is considered in the first vector, the rationality information determined by the first vector is more accurate.

[0078] In this embodiment, the first vector represents a first number of first trajectory points, and the rationality of the first predicted trajectory is determined based on the first vector, which greatly reduces the amount of computation and improves the efficiency of rationality judgment.

[0079] like Figure 4 As shown above, in the above Figure 1 Based on the illustrated embodiment, step 106 may include the following steps:

[0080] Step 1061: Using the second trajectory point as the base point, determine a first number of second trajectory points in the second predicted trajectory.

[0081] In some optional examples, the second trajectory point is used as the center trajectory point, and a second number of second trajectory points adjacent to the center trajectory point are determined in two directions in the second preset trajectory; the two sets of the second number of second trajectory points and the center trajectory point are used as the first number of second trajectory points.

[0082] In some alternative examples, a third number of second trajectory points adjacent to the starting trajectory point are determined along a second preset trajectory, with the second trajectory point as the starting trajectory point; the third number of second trajectory points and the starting trajectory point are used as the first number of second trajectory points.

[0083] Step 1062: Determine the second vector based on the first number of second trajectory points.

[0084] Optionally, the starting point and ending point of the second vector are determined in the order of the first number of second trajectory points in the second preset trajectory; the second vector is determined based on the starting point and ending point of the second vector.

[0085] In some optional embodiments, the process of determining the second vector may refer to the process of determining the first vector. In some optional examples, the corresponding first vector is determined for the i-th time of the second target b. or By determining the second vector, the second predicted trajectory can be decomposed into multiple second vectors. For example, when the second trajectory includes 90 second trajectory points, 90-2k or 90-j second vectors can be obtained. The number of second vectors is less than the number of second trajectory points. Under the premise that the number of first vectors is also less than the number of first trajectory points, the rationality information is determined by the first and second vectors, which further reduces the computational complexity and the amount of computation.

[0086] In the above embodiments, the larger the values ​​of k and j, the smaller the computational load, but the accuracy of the calculation results will be affected. Therefore, the values ​​of k and j are not necessarily better the larger they are. In this embodiment of the disclosure, through big data statistics and by calculating the yielding time between each other in the real trajectory of a large batch of interactive traffic participant data used for training / testing, that is, the time difference between the two traffic participants passing through the conflict point, it is determined that when the value of k is in the range of 2-5 or the value of j is in the range of 5-10, the accuracy of the rationality information can be improved while reducing the computational load.

[0087] like Figure 5a As shown above, in the above Figure 1 Based on the illustrated embodiment, step 108 may include the following steps:

[0088] Step 1081: Determine the positional relationship between at least one first vector and its corresponding second vector to obtain at least one positional relationship.

[0089] In this embodiment of the disclosure, each first vector corresponds to a first trajectory point at a time point, and each second vector corresponds to a second trajectory point at a time point. Furthermore, the first trajectory point and the second trajectory point have a corresponding relationship in the time dimension (both are multiple points predicted by the trajectory prediction method within the same time period in the future at the same time interval); therefore, the first vector and the second vector have a corresponding relationship based on the time dimension.

[0090] In this embodiment, the positional relationship between the first vector and its corresponding second vector includes intersection and non-intersection. Any method for calculating whether two vectors intersect can be used in this embodiment, and this embodiment does not limit the specific method for calculating the vector intersection. Furthermore, in the process of judging the reasonableness of the predicted trajectory, efficiency and accuracy are equally important. To speed up the calculation of whether two vectors intersect, the problem can be decomposed into whether the two endpoints of the first vector are respectively on both sides of the second vector, and whether the two endpoints of the second vector are respectively on both sides of the first vector. For example, as... Figure 5b As shown in the figure, the first vector corresponds to the first target a. The second vector corresponding to the second objective b There are intersections; reference Figure 5b As can be seen from the illustrated embodiment, it is possible to... (first vector) and The question of whether there is an intersection between (the second vector) is decomposed into: Are the two points respectively at? The two ends of the line formed, and Are the two points respectively at? The two ends of the straight line formed. And Are the two points respectively at? The two ends of the line formed can be calculated using the sign of the cross product of vectors. Therefore... and The intersection points between them can be expressed by the following formula (1):

[0091]

[0092] Wherein, when the above formula (1) holds true, it means and There are intersections between them; in addition, since each trajectory point includes x-axis coordinates and y-axis coordinates, each trajectory point is decomposed into corresponding x-axis coordinates and y-axis coordinates during the calculation process. At this time, formula (1) can be converted into the following formula (2) for expression, so as to realize the calculation:

[0093]

[0094] in,

[0095] Step 1082: Based on at least one positional relationship, determine the plausibility information of at least one of the first predicted trajectory and the second predicted trajectory.

[0096] This embodiment determines whether at least one of the first and second predicted trajectories is unreasonable by checking whether there is an intersection between the first and second vectors. Optionally, step 1082 may include:

[0097] Determine whether the positional relationship between the first vector and the corresponding second vector is a first positional relationship; a first positional relationship indicates that there is an intersection point between the first vector and the second vector.

[0098] In response to the existence of a first positional relationship in at least one positional relationship, the reasonableness information of at least one of the first predicted trajectory and the second predicted trajectory is determined to be unreasonable.

[0099] In this embodiment, when there is an intersection between any first vector and the corresponding second vector in at least one first vector, the reasonableness information is determined to be unreasonable. Only when there is no intersection between any first vector and the corresponding second vector is the reasonableness information determined to be reasonable. Correspondingly, step 1082 may also include: in response to the absence of a first positional relationship in at least one positional relationship, the reasonableness information of the first predicted trajectory and the second predicted trajectory is determined to be reasonable.

[0100] In this embodiment, a method is proposed to evaluate the rationality of the predicted trajectories of two targets by determining whether there is an intersection between the vectors determined by the trajectory points corresponding to the first and second targets. Furthermore, a method based on the similarity or difference of the signs of the vector cross product is used to quickly determine whether there is an intersection between the two vectors. The evaluation method provided in this embodiment can efficiently evaluate the rationality of the interaction trajectory in the time dimension and takes into account the yielding time. Applying this rationality information to predict the model can achieve the determination of trajectory prediction that is more in line with the actual application scenario (with reasonable yielding).

[0101] In some alternative embodiments, step 102 may include:

[0102] The trajectory of the first target is predicted by the trajectory prediction model, and the first predicted trajectory is obtained within a preset time period in the future.

[0103] The trajectory of the second target is predicted by the trajectory prediction model, and the second predicted trajectory is obtained within a preset time period in the future.

[0104] Optionally, the trajectory prediction model is trained.

[0105] In this embodiment, the order of prediction of the first target and the second target is not important. The first target can be predicted first, the second target can be predicted first, or the first target and the second target can be predicted simultaneously. In this embodiment, the trajectory prediction model can be any deep neural network model that can realize trajectory prediction in the prior art (e.g., VectorNet or Motion Transformer, etc.). This disclosure does not limit its specific network structure.

[0106] In some optional embodiments, the method provided in this embodiment may further include:

[0107] The trajectory prediction model is trained based on reasonable information.

[0108] Optionally, the training process may include: using a trajectory prediction model to predict the trajectories of the first training target and the second training target respectively, to obtain the first training predicted trajectory and the second training predicted trajectory.

[0109] Based on the above embodiments of this disclosure, the method for determining the rationality of predicted trajectories determines rationality information as the truth value, determines whether the first training predicted trajectory and / or the second training predicted trajectory are reasonable; adjusts the parameters of the trajectory prediction model according to the rationality information. For example, when the rationality information is unreasonable (e.g., 1 represents reasonable and 0 represents unreasonable), the rationality information and the probability value of the output trajectory are used to determine the loss value, and the network parameters of the trajectory prediction model are adjusted through the loss value (e.g., gradient backpropagation, etc.) to achieve training.

[0110] The trajectory prediction model trained in this embodiment can be the trajectory prediction model that determines the first and second predicted trajectories in the above embodiment, or it can be any network model that can achieve trajectory prediction. By using reasonableness information as supervision information, the trajectory prediction model not only achieves trajectory prediction, but also improves the reasonableness of the predicted trajectory.

[0111] Any method for determining the reasonableness of a predicted trajectory provided in this disclosure can be executed by any suitable device with data processing capabilities, including but not limited to: terminal devices and servers. Alternatively, any method for determining the reasonableness of a predicted trajectory provided in this disclosure can be executed by a processor, such as by a processor executing any method for determining the reasonableness of a predicted trajectory mentioned in this disclosure by calling corresponding instructions stored in memory. Further details will not be elaborated below.

[0112] Exemplary device

[0113] Figure 6 This is a schematic diagram of the structure of an apparatus for determining the reasonableness of a predicted trajectory provided in an exemplary embodiment of this disclosure. Figure 6As shown, the apparatus provided in this embodiment includes:

[0114] The trajectory prediction module 61 is used to obtain the first predicted trajectory of the first target within a preset time period in the future, and to obtain the second predicted trajectory of the second target within a preset time period in the future.

[0115] The first vector module 62 is used to determine at least one first vector corresponding to at least one first trajectory point in the first predicted trajectory.

[0116] The second vector module 63 is used to determine at least one second vector corresponding to at least one second trajectory point in the second predicted trajectory.

[0117] Reasonableness determination module 64 is used to determine reasonableness information of at least one of the first predicted trajectory and the second predicted trajectory based on at least one first vector and at least one second vector.

[0118] The apparatus for determining the rationality of a predicted trajectory provided in the above embodiments of this disclosure determines the rationality of the predicted trajectory corresponding to two targets by using vectors corresponding to two targets. Since the vectors combine multiple trajectory points, and each trajectory point corresponds to a time point, the rationality of the predicted trajectory is determined by combining the time dimension. Therefore, when determining the rationality of the predicted trajectory, this embodiment not only considers whether the two targets will collide and the time of collision, but also considers whether there is a reasonable buffer range for avoidance between the two targets. This allows for a more accurate and reasonable determination of the rationality of the predicted trajectory, facilitating more accurate and reasonable trajectory prediction in the future, and enabling safe driving control of the vehicle based on the predicted trajectory, thereby improving the safety of vehicle driving.

[0119] Figure 7 This is a schematic diagram of the structure of an apparatus for determining the reasonableness of a predicted trajectory, provided in another exemplary embodiment of this disclosure. Figure 7 As shown, the first vector module 62 includes:

[0120] The first trajectory point determination unit 621 is used to determine a first number of first trajectory points in the first predicted trajectory, using the first trajectory point as the base point.

[0121] The first trajectory vector unit 622 is used to determine the first vector based on a first number of first trajectory points.

[0122] Optionally, the first trajectory point determination unit 621 can be used to determine a second number of first trajectory points adjacent to the center trajectory point in two directions in a first preset trajectory, with the first trajectory point as the center trajectory point; and use the two sets of the second number of first trajectory points and the center trajectory point as the first number of first trajectory points.

[0123] Optionally, the first trajectory point determination unit 621 can also be used to determine a third number of first trajectory points adjacent to the starting trajectory point along a first preset trajectory, with the first trajectory point as the starting trajectory point; and to use the third number of first trajectory points and the starting trajectory point as the first number of first trajectory points.

[0124] Optionally, the first trajectory vector unit 622 is specifically used to determine the starting point and the ending point of the first vector in the order of the first number of first trajectory points in the first preset trajectory; and to determine the first vector based on the starting point and the ending point of the first vector.

[0125] like Figure 7 As shown, the second vector module 63 includes:

[0126] The second trajectory point determination unit 631 is used to determine a first number of second trajectory points in the second predicted trajectory, using the second trajectory point as the base point.

[0127] The second trajectory vector unit 632 is used to determine the second vector based on a first number of second trajectory points.

[0128] Optionally, the second trajectory point determination unit 631 can be used to determine a second number of second trajectory points adjacent to the center trajectory point in two directions in a second preset trajectory, with the second trajectory point as the center trajectory point; and use the two sets of the second number of second trajectory points and the center trajectory point as the first number of second trajectory points.

[0129] Optionally, the second trajectory point determination unit 631 can also be used to determine a third number of second trajectory points adjacent to the starting trajectory point along a second preset trajectory, using the second trajectory point as the starting trajectory point; and to use the third number of second trajectory points and the starting trajectory point as a first number of second trajectory points.

[0130] Optionally, the second trajectory vector unit 632 is specifically used to determine the starting point and ending point of the second vector in the order of the first number of second trajectory points in the second preset trajectory; and to determine the second vector based on the starting point and ending point of the second vector.

[0131] Figure 8 This is a schematic diagram of the structure of an apparatus for determining the reasonableness of a predicted trajectory, provided in yet another exemplary embodiment of this disclosure. Figure 8 As shown, the rationality determination module 64 includes:

[0132] The positional relationship determination unit 641 is used to determine the positional relationship between at least one first vector and the corresponding second vector respectively, and to obtain at least one of the positional relationships; the first vector and the second vector have a corresponding relationship based on the time dimension.

[0133] Information determination unit 642 is used to determine the rationality information of at least one of the first predicted trajectory and the second predicted trajectory based on at least one positional relationship.

[0134] Optionally, the information determining unit 642 is specifically used to determine whether the positional relationship between the first vector and the corresponding second vector is a first positional relationship; the first positional relationship indicates that there is an intersection point between the first vector and the second vector; in response to the existence of a first positional relationship in at least one positional relationship, the reasonableness information of at least one of the first predicted trajectory and the second predicted trajectory is determined to be unreasonable.

[0135] In some optional embodiments, the trajectory prediction module 61 is specifically used to predict the trajectory of the first target using a trajectory prediction model to obtain a first predicted trajectory within a preset time period in the future; and to predict the trajectory of the second target using the trajectory prediction model to obtain a second predicted trajectory within a preset time period in the future.

[0136] In some optional embodiments, the apparatus provided in this disclosure may further include:

[0137] The model training module (not shown in the figure) is used to train the trajectory prediction model based on rationality information.

[0138] The exemplary embodiments of this device and the exemplary embodiments of the above methods can be implemented with reference to each other. The beneficial technical effects corresponding to the exemplary embodiments of this device can be found in the corresponding beneficial technical effects of the exemplary method section above, and will not be repeated here.

[0139] Exemplary electronic devices

[0140] Below, for reference Figure 9 This describes an electronic device according to embodiments of the present disclosure. The electronic device may be either or both of a first device and a second device, or a standalone device independent of them, which may communicate with the first device and the second device to receive acquired input signals from them.

[0141] Figure 9 A block diagram of an electronic device according to an embodiment of the present disclosure is shown.

[0142] like Figure 9 As shown, the electronic device includes one or more processors and memory.

[0143] A processor can be a central processing unit (CPU) or other form of processing unit with data processing and / or instruction execution capabilities, and can control other components in an electronic device to perform desired functions.

[0144] The memory can store one or more computer program products, and the memory can include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may include, for example, random access memory (RAM) and / or cache memory. The non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc. One or more computer program products can be stored on the computer-readable storage medium, and the processor can run the computer program products to implement the methods for determining the reasonableness of predicted trajectories in the various embodiments of this disclosure described above, and / or other desired functions.

[0145] In one example, the electronic device may also include input devices and output devices, which are interconnected via a bus system and / or other forms of connection mechanism (not shown).

[0146] In addition, the input device may also include, for example, a keyboard, a mouse, etc.

[0147] This output device can output various information to the outside, including determined distance information, direction information, etc. The output device may include, for example, a display, a speaker, a printer, and a communication network and its connected remote output devices, etc.

[0148] Of course, for the sake of simplicity, Figure 9 Only some of the components of the electronic device relevant to this disclosure are shown, omitting components such as buses, input / output interfaces, etc. In addition, the electronic device may include any other suitable components depending on the specific application.

[0149] In addition to the methods and apparatus described above, embodiments of this disclosure may also be computer program products comprising computer program instructions that, when executed by a processor, cause the processor to perform the steps in the methods for determining the reasonableness of a predicted trajectory according to various embodiments of this disclosure as described in the foregoing portion of this specification.

[0150] The computer program product can be written in any combination of one or more programming languages ​​to perform the operations of the embodiments of this disclosure. The programming languages ​​include object-oriented programming languages ​​such as Java and C++, as well as conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on a user's computing device, partially on a user's computing device, as a standalone software package, partially on a user's computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0151] Furthermore, embodiments of this disclosure may also be computer-readable storage media storing computer program instructions thereon, which, when executed by a processor, cause the processor to perform the steps in the methods for determining the reasonableness of a predicted trajectory according to various embodiments of this disclosure as described in the foregoing portion of this specification.

[0152] The computer-readable storage medium may be any combination of one or more readable media. A readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may, for example, include, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatuses, or devices, or any combination thereof. More specific examples of readable storage media (a non-exhaustive list) include: electrical connections having one or more wires, portable disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0153] The basic principles of this disclosure have been described above with reference to specific embodiments. However, it should be noted that the advantages, benefits, and effects mentioned in this disclosure are merely examples and not limitations, and should not be considered as essential features of each embodiment of this disclosure. Furthermore, the specific details disclosed above are for illustrative and facilitative purposes only, and are not limitations. These details do not limit the scope of this disclosure to the necessity of employing the aforementioned specific details for implementation.

[0154] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For system embodiments, since they largely correspond to method embodiments, the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments.

[0155] The block diagrams of devices, apparatuses, devices, and systems disclosed herein are merely illustrative examples and are not intended to require or imply that they must be connected, arranged, or configured in the manner shown in the block diagrams. As those skilled in the art will recognize, these devices, apparatuses, devices, and systems can be connected, arranged, and configured in any manner. Words such as “comprising,” “including,” “having,” etc., are open-ended terms meaning “including but not limited to,” and are used interchangeably with them. The terms “or” and “and” as used herein refer to the terms “and / or,” and are used interchangeably with them unless the context clearly indicates otherwise. The term “such as” as used herein refers to the phrase “such as but not limited to,” and is used interchangeably with it.

[0156] The methods and apparatus of this disclosure may be implemented in many ways. For example, they may be implemented by software, hardware, firmware, or any combination of software, hardware, and firmware. The above-described order of steps for the methods is for illustrative purposes only, and the steps of the methods of this disclosure are not limited to the order specifically described above unless otherwise specifically stated. Furthermore, in some embodiments, this disclosure may also be implemented as a program recorded on a recording medium, the program including machine-readable instructions for implementing the methods according to this disclosure. Thus, this disclosure also covers recording media storing programs for performing the methods according to this disclosure.

[0157] It should also be noted that in the apparatus, devices, and methods of this disclosure, the components or steps can be disassembled and / or recombined. These disassemblies and / or recombinations should be considered as equivalent solutions to this disclosure.

[0158] The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use this disclosure. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to other aspects without departing from the scope of this disclosure. Therefore, this disclosure is not intended to be limited to the aspects shown herein, but rather to be carried out within the widest scope consistent with the principles and novel features disclosed herein.

[0159] The above description has been given for purposes of illustration and description. Furthermore, this description is not intended to limit the embodiments of this disclosure to the forms disclosed herein. Although numerous exemplary aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, alterations, additions, and sub-combinations therein.

Claims

1. A method for determining the rationality of a predicted trajectory, characterized in that, include: Obtain the first predicted trajectory of the first target within a future preset time period, and obtain the second predicted trajectory of the second target within the same future preset time period; Determine at least one first vector corresponding to at least one first trajectory point in the first predicted trajectory; Determine at least one second vector corresponding to at least one second trajectory point in the second predicted trajectory; Based on the at least one first vector and the at least one second vector, determine the plausibility information of at least one of the first predicted trajectory and the second predicted trajectory.

2. The method according to claim 1, characterized in that, Determining at least one first vector corresponding to at least one first trajectory point in the first predicted trajectory includes: Using the first trajectory point as a base point, determine a first number of the first trajectory points in the first predicted trajectory; The first vector is determined based on the first number of the first trajectory points.

3. The method according to claim 2, characterized in that, The step of determining a first number of the first trajectory points in the first predicted trajectory, using the first trajectory point as a base point, includes: Taking the first trajectory point as the center trajectory point, determine a second number of first trajectory points adjacent to the center trajectory point in two directions in the first preset trajectory; The first trajectory points of the first number are used as the first number of the two sets of the second quantity and the center trajectory point.

4. The method according to claim 2, characterized in that, The step of determining a first number of the first trajectory points in the first predicted trajectory, using the first trajectory point as a base point, includes: Using the first trajectory point as the starting trajectory point, determine a third number of first trajectory points adjacent to the starting trajectory point along the first preset trajectory; The third number of first trajectory points and the starting trajectory point are used as the first number of first trajectory points.

5. The method according to any one of claims 2-4, characterized in that, Determining the first vector based on the first number of the first trajectory points includes: The starting point and ending point of the first vector are determined according to the order of the first number of the first trajectory points in the first preset trajectory; The first vector is determined based on the starting point and the ending point of the first vector.

6. The method according to any one of claims 1-5, characterized in that, Determining at least one second vector corresponding to at least one second trajectory point in the second predicted trajectory includes: Using the second trajectory point as a base point, a first number of the second trajectory points are determined in the second predicted trajectory; The second vector is determined based on the first number of second trajectory points.

7. The method according to claim 6, characterized in that, The step of determining a first number of second trajectory points in the second predicted trajectory, using the second trajectory point as a base point, includes: Using the second trajectory point as the center trajectory point, determine a second number of second trajectory points adjacent to the center trajectory point in two directions in the second preset trajectory; The second number of second trajectory points and the center trajectory point are used as the first number of second trajectory points.

8. The method according to claim 6, characterized in that, The step of determining a first number of second trajectory points in the second predicted trajectory, using the second trajectory point as a base point, includes: Using the second trajectory point as the starting trajectory point, determine a third number of second trajectory points adjacent to the starting trajectory point along the second preset trajectory; The third number of second trajectory points and the starting trajectory point are used as the first number of second trajectory points.

9. The method according to any one of claims 6-8, characterized in that, Determining the second vector based on the first number of second trajectory points includes: The starting point and ending point of the second vector are determined according to the order of the first number of second trajectory points in the second preset trajectory; The second vector is determined based on the starting point and the ending point of the second vector.

10. The method according to any one of claims 1-9, characterized in that, The step of determining the plausibility information of at least one of the first predicted trajectory and the second predicted trajectory based on the at least one first vector and the at least one second vector includes: Determine the positional relationship between at least one of the first vectors and the corresponding second vectors to obtain at least one such positional relationship; the first vector and the second vector have a correspondence based on the time dimension. Based on at least one of the said positional relationships, determine the plausibility information of at least one of the first predicted trajectory and the second predicted trajectory.

11. The method according to claim 10, characterized in that, The step of determining the plausibility information of at least one of the first predicted trajectory and the second predicted trajectory based on at least one of the said positional relationships includes: Determine whether the positional relationship between the first vector and the corresponding second vector is a first positional relationship; the first positional relationship indicates that there is an intersection point between the first vector and the second vector. In response to the existence of the first positional relationship in at least one of the positional relationships, the reasonableness information of at least one of the first predicted trajectory and the second predicted trajectory is determined to be unreasonable.

12. The method according to any one of claims 1-11, characterized in that, The steps of obtaining the first predicted trajectory of the first target within a future preset time period and obtaining the second predicted trajectory of the second target within the future preset time period include: The trajectory of the first target is predicted by the trajectory prediction model to obtain the first predicted trajectory within the future preset time period. The trajectory prediction model is used to predict the trajectory of the second target, thereby obtaining the second predicted trajectory within the preset future time period.

13. The method according to any one of claims 1-12, characterized in that, Also includes: The trajectory prediction model is trained based on the aforementioned rationality information.

14. An apparatus for determining the rationality of a predicted trajectory, characterized in that, include: The trajectory prediction module is used to obtain a first predicted trajectory of a first target within a future preset time period, and to obtain a second predicted trajectory of a second target within the same future preset time period. The first vector module is used to determine at least one second vector corresponding to at least one second trajectory point in the second predicted trajectory; The second vector module is used to determine at least one second vector corresponding to at least one second trajectory point in the second predicted trajectory; A rationality determination module is used to determine the rationality information of at least one of the first predicted trajectory and the second predicted trajectory based on the at least one first vector and the at least one second vector.

15. An electronic device, characterized in that, include: Memory, used to store computer program products; A processor is configured to execute a computer program product stored in the memory, wherein, when the computer program product is executed, it implements the method for determining the reasonableness of a predicted trajectory as described in any one of claims 1-13.

16. A computer-readable storage medium having computer program instructions stored thereon, characterized in that, When the computer program instructions are executed by the processor, they implement the method for determining the reasonableness of the predicted trajectory as described in any one of claims 1-13.

17. A computer program product comprising computer program instructions, characterized in that, When the computer program instructions are executed by the processor, they implement the method for determining the reasonableness of the predicted trajectory as described in any one of claims 1-13.

Images