A method, apparatus, device, and autonomous vehicle for correcting a driving trajectory

Abstract

The present disclosure provides a method, apparatus, device, and autonomous vehicle for correcting a driving trajectory, relating to the field of artificial intelligence technology, and particularly to the fields of autonomous driving, intelligent transportation, and deep learning technology. The specific implementation solution is as follows: obtaining obstacle boundary information of a road section where a predicted driving trajectory of a driving object is located; wherein the obstacle boundary information at least includes position information of obstacle points in the obstacle boundary; correcting the predicted driving trajectory according to the position information of the obstacle points in the obstacle boundary, the position information of candidate trajectory points in the predicted driving trajectory, and the lateral safety distance of the driving object when driving on the road section. It can correct the error of the predicted driving trajectory, improve the accuracy of the predicted driving trajectory, and further improve the safety of autonomous driving.

Description

Technical Field

[0001] The present disclosure relates to the field of artificial intelligence technology, and more particularly to the fields of autonomous driving, intelligent transportation, and deep learning technology. Specifically, it relates to a method, device, equipment, and autonomous driving vehicle for correcting a driving trajectory. Background Art

[0002] With the development of artificial intelligence technology, autonomous driving technology has gradually emerged. And the prediction of the driving trajectory is a core link in the process of autonomous driving. For example, predicting the driving trajectory of the current vehicle, and predicting the driving trajectories of dynamic obstacles around the current vehicle, etc. However, the predicted driving trajectory may have errors, seriously affecting the safety of autonomous driving, and there is an urgent need for improvement. Summary of the Invention

[0003] The present disclosure provides a method, device, equipment, and autonomous driving vehicle for correcting a driving trajectory.

[0004] According to one aspect of the present disclosure, there is provided a method for correcting a driving trajectory, including:

[0005] Obtaining obstacle boundary information of a section where the predicted driving trajectory of a driving object is located; wherein the obstacle boundary information at least includes position information of obstacle points in the obstacle boundary;

[0006] Correcting the predicted driving trajectory according to the position information of the obstacle points in the obstacle boundary, the position information of candidate trajectory points in the predicted driving trajectory, and the lateral safety distance of the driving object when driving in the section.

[0007] According to another aspect of the present disclosure, there is provided an electronic device, which includes:

[0008] At least one processor; and

[0009] A memory communicatively connected to at least one processor; wherein,

[0010] The memory stores instructions executable by at least one processor, and the instructions are executed by at least one processor to enable at least one processor to execute the method for correcting a driving trajectory according to any embodiment of the present disclosure.

[0011] According to another aspect of the present disclosure, there is provided a non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to cause a computer to execute the method for correcting a driving trajectory according to any embodiment of the present disclosure.

[0012] According to another aspect of the present disclosure, there is provided an autonomous driving vehicle, including the electronic device according to the embodiment of the present disclosure.

[0013] The solution of the embodiment of the present disclosure can correct the error of the predicted driving trajectory, improve the accuracy of the predicted driving trajectory, and further improve the safety of autonomous driving.

[0014] It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it used to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The drawings are used to better understand the solution of the present disclosure and do not constitute a limitation to the present disclosure. Among them:

[0016] Figure 1 is a flowchart of a method for correcting a driving trajectory provided according to an embodiment of the present disclosure;

[0017] Figure 2A is a flowchart of a method for correcting a driving trajectory provided according to an embodiment of the present disclosure;

[0018] Figure 2B is a schematic diagram of the principle for correcting the position of a trajectory point provided according to an embodiment of the present disclosure;

[0019] Figure 3A is a flowchart of a method for correcting a driving trajectory provided according to an embodiment of the present disclosure;

[0020] Figure 3B is a schematic diagram of the principle for adding missing trajectory points provided according to an embodiment of the present disclosure;

[0021] Figure 4A is a flowchart of a method for correcting a driving trajectory provided according to an embodiment of the present disclosure;

[0022] Figure 4B is a schematic diagram of the principle for screening missing trajectory points provided according to an embodiment of the present disclosure;

[0023] Figure 5 is a flowchart of a method for correcting a driving trajectory provided according to an embodiment of the present disclosure;

[0024] Figure 6 is a schematic diagram of the structure of a device for correcting a driving trajectory provided according to an embodiment of the present disclosure;

[0025] Figure 7 is a block diagram of an electronic device for implementing the method for correcting a driving trajectory of the embodiment of the present disclosure. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0026] The exemplary embodiments of the present disclosure will be described below in conjunction with the accompanying drawings. Various details of the embodiments of the present disclosure are included to facilitate understanding, and they should be considered merely exemplary. Therefore, those of ordinary skill in the art should recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, descriptions of well-known functions and structures are omitted in the following description for clarity and conciseness.

[0027] Figure 1 FIG. is a flowchart of a method for correcting a driving trajectory provided according to an embodiment of the present disclosure; the embodiments of the present disclosure are applicable to the situation of correcting a predicted driving trajectory. In particular, it is applicable to the situation of correcting a driving trajectory predicted by a deep learning model for a driving object. This method can be executed by a device for correcting a driving trajectory, and the device can be implemented in software and / or hardware. Specifically, it can be integrated into an electronic device with a driving trajectory prediction or use function. As Figure 1 shown, the method for correcting a driving trajectory provided in this embodiment may include:

[0028] S101, obtaining obstacle boundary information of the section where the predicted driving trajectory of the driving object is located.

[0029] Among them, the driving object in this embodiment can be any driving object on the road. For example, it can be a vehicle and a pedestrian on the road, etc. For example, in an autonomous driving scenario, the driving object can be an autonomous vehicle, or a dynamic obstacle such as a pedestrian or other vehicle around the autonomous vehicle.

[0030] The predicted driving trajectory can be a driving trajectory predicted for the driving object in a certain way. For example, it can be a driving trajectory predicted for the driving object by a deep learning model; it can also be a section of driving trajectory simulated for the driving object by a trajectory simulator; it can also be a section of driving trajectory planned for the driving object according to a route planning strategy, etc. It should be noted that the predicted driving trajectory is composed of multiple trajectory points.

[0031] The obstacle boundary can be a real object object (i.e., the hard boundary on both sides of the road) used to represent the road boundary on both sides of the road, which corresponds to the virtual boundary (i.e., the soft boundary) on both sides of the road. For example, the curbs, fences or cone barrels at the boundary positions on both sides of the road are the obstacle boundaries (i.e., the hard boundaries) of the road; the road lines at the boundary positions on both sides of the road are the virtual boundaries (i.e., the soft boundaries) of the road.

[0032] The obstacle boundary information may be information describing the attributes related to the obstacle boundary, which at least includes the position information of the obstacle points in the obstacle boundary. Among them, the obstacle points in the obstacle boundary may be several position points where the obstacle boundary intersects the ground. For example, if the obstacle boundary is formed by arranging several obstacles (such as cone barrels) in sequence, then the intersection point of each obstacle (such as a cone barrel) and the ground can be used as an obstacle point of the obstacle boundary. If the obstacle boundary is formed by a continuous plane, such as a curb, then at this time, several intersection points with the ground can be extracted from this continuous plane as the obstacle points of the obstacle boundary. The extraction rule of the obstacle points is not limited in this embodiment. For example, one point can be extracted as an obstacle point at a preset distance interval. Optionally, in this embodiment, the position information of the obstacle points may be the lateral position coordinate and the longitudinal position coordinate of the obstacle points relative to a reference point (such as the starting point of the center lane line of the road section) in the road section where they are located.

[0033] Optionally, the obstacle boundary information in this embodiment may further include other attribute information of the obstacle boundary, such as the obstacle boundary type. Among them, the obstacle boundary type may include but is not limited to: curbs, fences, or cone barrels, etc.

[0034] Optionally, in this embodiment, the predicted driving trajectory of the driving object may be obtained first, and then based on the position information of the trajectory points in the predicted driving trajectory and the high-precision map, at least one road segment (i.e., the road section to which it belongs) to which the predicted driving trajectory belongs is determined, and then it is sequentially determined whether there is an obstacle boundary in each road section. If so, the obstacle boundary information corresponding to the road section is obtained; otherwise, the operation of obtaining the obstacle boundary is not performed.

[0035] Optionally, in this embodiment, there are many ways to determine whether there is an obstacle boundary in a road section and obtain the obstacle boundary information, which is not limited herein. Specifically, one implementable way is to analyze whether there is an obstacle boundary in the road section through the road panoramic view corresponding to the road section in the high-precision map or the real-time scene map of the road section. If so, further analyze and obtain the corresponding obstacle boundary information.

[0036] Another implementation method is as follows: First, construct an obstacle boundary map that contains the obstacle boundary information of the global road segments in advance. The obstacle boundary map should at least include the road segments associated with the obstacle boundary in the high-precision map, and mark the obstacle boundary information corresponding to each road segment in the obstacle boundary map in a certain way. For example, the obstacle boundary information can be added to each road segment in the form of key-value pairs. Here, the key is the identifier of the road segment to be marked (such as road name or number, etc.), and the value is the obstacle boundary information corresponding to the road segment. At this time, in this embodiment, based on the road identifier of the road segment where the predicted driving trajectory is located, it is possible to check whether there is an associated obstacle boundary in the pre-constructed obstacle boundary map. If so, the hard boundary information corresponding to this road segment can be further obtained from the obstacle boundary map.

[0037] S102. According to the position information of the obstacle points in the obstacle boundary, the position information of the candidate trajectory points in the predicted driving trajectory, and the lateral safety distance of the driving object when driving on the road segment, correct the predicted driving trajectory.

[0038] Among them, the candidate trajectory points in the predicted driving trajectory can be each trajectory point in the predicted driving trajectory, or some trajectory points extracted from the trajectory points in the driving trajectory according to a certain strategy. This embodiment does not limit the specific extraction strategy. For example, a candidate trajectory point can be extracted every preset time period, or a candidate trajectory point can be extracted every preset distance; it can also be that a candidate trajectory point is extracted every preset number of trajectory points, etc. It should be noted that in order to improve the smoothness of the trajectory correction result and reduce the complexity in the correction process, this embodiment preferably selects some trajectory points taken from the driving trajectory as candidate trajectory points.

[0039] The lateral safety distance can be the minimum lateral driving distance from the obstacle boundary set in advance for the driving object when driving on this road segment to ensure the driving safety of the driving object. Optionally, in this embodiment, the size of the lateral safety distance can be a preset fixed distance value, or it can be adjusted in real time according to different driving objects, road segments where the driving object is located, and obstacle boundaries according to a certain strategy, and this is not limited here.

[0040] Optionally, an implementation of this embodiment is as follows: Based on the position information of the obstacle points in the obstacle boundary and the position information of the candidate trajectory points in the predicted driving trajectory, determine the relative positions between the candidate trajectory points and the obstacle points. Then, analyze whether there are any unreasonable situations among the candidate trajectory points of the predicted driving trajectory based on the relative positions. For example, there may be situations where the distance from the obstacle boundary on both sides of the road is too close or the obstacle boundary is crossed. If such situations exist, the position coordinates of the unreasonably located trajectory points can be corrected based on the position information of the obstacle points, the predicted position information of the unreasonably located candidate trajectory points, and the lateral safety distance of the driving object when driving on this section of the road, and / or by adding new trajectory points to the predicted driving trajectory to ensure the rationality of each trajectory point in the predicted driving trajectory, etc.

[0041] Another implementation of this embodiment is to input the position information of the candidate trajectory points in the predicted driving trajectory, the position information of the obstacle points in the obstacle boundary of the section where the predicted driving trajectory is located, and the lateral safety distance of the driving object when driving on this section of the road into a pre-trained trajectory correction model. The model can then analyze the unreasonably located trajectory points in the driving trajectory based on the input data and output the corrected position information corresponding to the unreasonably located trajectory points.

[0042] The solution of the embodiment of the present disclosure obtains the position information of the obstacle points in the obstacle boundary of the section where the predicted driving trajectory of the driving object is located, and corrects the predicted driving trajectory based on the position information of the obstacle points, the position information of the candidate trajectory points in the predicted driving trajectory, and the lateral safety distance of the driving object when driving on the section of the road. This solution can correct the error of the predicted driving trajectory, and during the process of correcting the predicted driving trajectory, it focuses on considering the obstacle boundary information of the section, effectively reducing unreasonable problems such as the predicted trajectory being too close to the obstacle boundary or crossing the obstacle boundary, improving the accuracy of the predicted driving trajectory, and thus improving the safety of autonomous driving.

[0043] Optionally, in this embodiment, a preferred method for determining the lateral safety distance of the driving object when driving on the section of the road is as follows: Determine the driving width of the driving object according to the type of the driving object; determine the lateral safety distance of the driving object when driving on this section of the road according to the driving width and the buffer distance.

[0044] Specifically, in this embodiment, the width of the moving object can be determined according to the type of the moving object, and the width of the moving object can be used as the driving width. For example, if the moving object is a vehicle, the dimension information of the vehicle of this type can be searched according to the vehicle type, and then the vehicle width of the vehicle can be determined. Then, the driving width of the moving object is directly added to the buffer distance, or multiplied by a preset multiple (such as 0.5 times) and then added to the buffer distance to obtain the lateral safety distance of the moving object when driving on this section of the road. This solution determines a dedicated lateral safety distance for different moving objects according to their types, improving the accuracy and rationality of determining the lateral safety distance.

[0045] Among them, the buffer distance in this solution can be a fixed value set in advance according to experience, or determined according to certain rules in combination with the actual scenario. Optionally, if the buffer distance in this embodiment is determined in real time according to certain rules, the determination method can be: determining the buffer distance according to the road width of the section, the obstacle boundary type in the obstacle boundary information, and the type of the moving object. Specifically, the safety distance range to be reserved around the obstacle boundary can be determined according to the obstacle boundary type in the obstacle boundary information; the driving width of the moving object can be determined according to the type of the moving object, and then, in combination with the actual road width, the safety distance range of the current section, and the constraint relationship among the driving widths of the moving objects, the optimal buffer distance can be given. This solution determines a dedicated buffer distance for the moving object by considering the road width, the obstacle boundary type, and the type of the moving object, improving the accuracy of determining the buffer distance, and thus ensuring the accuracy of determining the lateral safety distance.

[0046] Figure 2A is a flowchart of a method for correcting a driving trajectory according to an embodiment of the present disclosure; Figure 2B is a schematic diagram of the principle of correcting the position of a trajectory point according to an embodiment of the present disclosure. On the basis of the above embodiment, the embodiment of the present disclosure further explains in detail how to correct the predicted driving trajectory according to the position information of the obstacle points in the obstacle boundary, the position information of the candidate trajectory points in the predicted driving trajectory, and the lateral safety distance of the moving object when driving on the section of the road. As Figure 2A-2B shown, the method for correcting a driving trajectory provided in this embodiment may include:

[0047] S201, obtaining the obstacle boundary information of the section where the predicted driving trajectory of the moving object is located.

[0048] Among them, the obstacle boundary information at least includes the position information of the obstacle points in the obstacle boundary.

[0049] S202. Determine the error trajectory points among the candidate trajectory points according to the position information of the obstacle points in the obstacle boundary, the position information of the candidate trajectory points in the predicted driving trajectory, and the lateral safety distance of the driving object when driving on this section of the road.

[0050] Among them, the error trajectory points in this embodiment may be the trajectory points in the driving trajectory that have position errors due to being too close to the obstacle boundary. Preferably, in this embodiment, the trajectory points with a distance from the obstacle boundary less than the lateral safety distance may be used as the error trajectory points.

[0051] Optionally, an implementable manner in this embodiment is: input the position information of the obstacle points in the obstacle boundary, the position information of each candidate trajectory point in the predicted driving trajectory, and the lateral safety distance of the driving object when driving on this section of the road into the error analysis model, and this model can analyze whether the candidate trajectory point is an error trajectory point based on the data information.

[0052] Another implementable manner is: determine the lateral predicted distance between the candidate trajectory point and the obstacle boundary according to the position information of the obstacle points in the obstacle boundary and the position information of the candidate trajectory points in the predicted driving trajectory; if the lateral predicted distance is less than the lateral safety distance of the driving object when driving on the section of the road, then use the candidate trajectory point as the error trajectory point. Specifically, for each candidate trajectory point in the predicted driving trajectory, an associated obstacle point can be determined for it from each obstacle point of the obstacle boundary. And by combining the position information of this candidate trajectory point and the position information of its associated obstacle point, calculate the lateral predicted distance between the candidate trajectory point and the obstacle boundary. Then judge whether the lateral predicted distance is less than the lateral safety distance of the driving object when driving on the section of the road. If so, then use this candidate trajectory point as the error trajectory point. This embodiment preferably adopts this implementable manner to determine the error trajectory points. This manner can achieve a more accurate and comprehensive determination of the error trajectory points, providing a guarantee for the subsequent accurate and comprehensive correction of the predicted driving trajectory.

[0053] Exemplarily, such as Figure 2BAs shown, b1 - b3 are three obstacle points of the obstacle boundary, and p1 and p2 are two candidate trajectory points in the predicted driving trajectory. One implementable way in this embodiment is: taking the obstacle point closest to the candidate trajectory point as the obstacle point associated with the candidate trajectory point. For example, taking the obstacle point b1 as the obstacle point associated with the candidate trajectory point p1; taking the obstacle point b2 as the obstacle point associated with the candidate trajectory point p2. At this time, the difference between the lateral position coordinate of the candidate trajectory point and the lateral position coordinate of its associated obstacle point can be calculated as the lateral predicted distance between the candidate trajectory point and the obstacle boundary. Another implementable way is: judging whether there is an obstacle point with the same ordinate as the candidate trajectory point among the obstacle points of the obstacle boundary. If so, taking it as the obstacle point associated with the candidate trajectory point; otherwise, taking the two closest obstacle points in front of and behind the candidate trajectory point as the obstacle points associated with the candidate trajectory point. For example, the obstacle point b1 can be taken as the obstacle point associated with the candidate trajectory point p1; the obstacle points b2 and b3 can be taken as the obstacle points associated with the candidate trajectory point p2. At this time, the difference between the lateral position coordinate of the candidate trajectory point and the lateral position coordinates of its associated obstacle points can be calculated first. If the number of obstacle points associated with the candidate trajectory point is one, directly taking this difference as the lateral predicted distance between the candidate trajectory point and the obstacle boundary; if the number of obstacle points associated with the candidate trajectory point is multiple, the difference between the lateral position coordinates of the candidate trajectory point and each obstacle point can be averaged or weighted averaged to obtain the lateral predicted distance between the candidate trajectory point and the obstacle boundary. Among them, if weighted averaging is performed, the weight can be determined according to the difference between the longitudinal coordinates of the candidate trajectory point and its associated obstacle points. For example, the greater the longitudinal distance difference, the smaller the weight.

[0054] S203. Perform position correction on the error trajectory points in the predicted driving trajectory according to the lateral safety distance.

[0055] Optionally, in this embodiment, the lateral position coordinate of the error trajectory point in the predicted driving trajectory can be corrected to this lateral safety distance. The longitudinal position coordinate of the error trajectory point can be not adjusted, or can be adaptively adjusted according to certain rules, which is not limited in this embodiment.

[0056] Exemplarily, as Figure 2B shown, correcting the error trajectory point p1 to the position of p1'; correcting the error trajectory point p2 to the position of p2'.

[0057] It should be noted that this embodiment mainly corrects the situation where the distance to the obstacle boundary in the predicted driving trajectory is too close.

[0058] The solution of the embodiment of the present disclosure obtains the position information of the obstacle points in the obstacle boundary of the section where the predicted driving trajectory of the driving object is located, and based on the position information of the obstacle points, the position information of the candidate trajectory points in the predicted driving trajectory, and the lateral safety distance of the driving object when driving on the section, first determines the error trajectory points, and corrects the error trajectory points based on the lateral safety distance to realize the correction of the predicted driving trajectory. According to the position information of the obstacle points, the position information of the candidate trajectory points, and the lateral safety distance, this solution can accurately and quickly screen out the unreasonable estimated points in the candidate trajectory points that are too close to the obstacle boundary and correct them, so that the corrected predicted driving trajectory will not be too close to the obstacle boundary, greatly improving the accuracy of the predicted driving trajectory.

[0059] Figure 3A is a flowchart of a method for correcting a driving trajectory provided according to an embodiment of the present disclosure; Figure 3B is a schematic diagram of the principle of adding missing trajectory points provided according to an embodiment of the present disclosure. On the basis of the above embodiment, the embodiment of the present disclosure further explains in detail how to correct the predicted driving trajectory according to the position information of the obstacle points in the obstacle boundary, the position information of the candidate trajectory points in the predicted driving trajectory, and the lateral safety distance of the driving object when driving on the section, as Figure 3A-3B shown, the method for correcting a driving trajectory provided in this embodiment may include:

[0060] S301, obtain the obstacle boundary information of the section where the predicted driving trajectory of the driving object is located.

[0061] Among them, the obstacle boundary information at least includes the position information of the obstacle points in the obstacle boundary.

[0062] S302, according to the position information of the obstacle points in the obstacle boundary and the position information of the candidate trajectory points in the predicted driving trajectory, determine the target trajectory point pairs in the predicted driving trajectory, and the target points included between the target trajectory point pairs.

[0063] In this embodiment, if an obstacle point of the obstacle boundary is included between two adjacent candidate trajectory points in the driving trajectory, then these two adjacent candidate trajectory points are used as a group of target trajectory point pairs. And the obstacle point located in this group of target trajectory point pairs is used as the target point, that is, the target point belongs to the obstacle point.

[0064] Optionally, in this embodiment, the target trajectory point pairs and the target points included in the target trajectory point pairs can be determined by judging the relationship between the longitudinal position coordinates of the adjacent candidate trajectory points in the driving trajectory and the longitudinal position coordinates of the obstacle points of the obstacle boundary.

[0065] Specifically, in this embodiment, for two adjacent trajectory points in the driving trajectory in sequence, it is determined whether the longitudinal position coordinates of one or more obstacle points are covered between the longitudinal position coordinates of these two adjacent trajectory points. If so, these two adjacent trajectory points are used as a pair of target trajectory points, and the obstacle points whose longitudinal position coordinates are between the target trajectory point pair are used as the target points included in this target trajectory point pair.

[0066] Exemplarily, as Figure 3B shown, b4 - b6 are three obstacle points of the obstacle boundary, and p3 and p4 are two adjacent candidate trajectory points in the predicted driving trajectory; it can be seen from Figure 3B that the obstacle points b4 - b6 are all located between the candidate trajectory point p3 and the candidate trajectory point p4. Then, the candidate trajectory point p3 and the candidate trajectory point p4 are used as a pair of target trajectory points, and the obstacle points b4 - b6 are used as the target points included in this group of target trajectory points.

[0067] S303. Determine the position information of the missing trajectory point according to the position information of the target point and the lateral safety distance of the driving object when driving on the road section.

[0068] Optionally, an implementable manner of this embodiment is: taking the longitudinal position coordinate of the target trajectory point as the longitudinal position coordinate of the missing trajectory point, and taking the lateral safety distance of the driving object when driving on the road section as the lateral position coordinate of the missing trajectory point, thus obtaining the position information of the missing trajectory point.

[0069] If the number of target points is multiple, the first implementable manner can be adopted to calculate a missing trajectory point for each target point, or the second implementable manner can be adopted below to determine a missing trajectory point based on multiple target points. The specific implementation manner is: determine the risk coefficient of the target point according to the position information of the target point; specifically, the risk coefficient of each target point can be calculated through the following formula (1).

[0070] R = 1 / [α * l 目标 + (1 - α) * s 目标 (1)

[0071] where R is the risk coefficient of the target point; α is a preset calculation parameter, and its value is greater than or equal to 0 and less than or equal to 1. For example, it can be taken as 0.99. l 目标 is the lateral position coordinate of the target point; s 目标 is the longitudinal position coordinate of the target point.

[0072] After determining the risk coefficient of the target point, the dangerous point is determined from the target points according to the risk coefficient of the target point; for example, the target point with the highest risk coefficient may be selected as the dangerous point; and then the position information of the missing track point is determined according to the position information of the dangerous point and the lateral safety distance of the driving object on the road section. For example, the longitudinal position coordinate of the dangerous point may be used as the longitudinal position coordinate of the missing track point, and the lateral safety distance of the driving object on the road section may be used as the lateral position coordinate of the missing track point.

[0073] For example, Figure 3B As shown, assuming that the risk coefficient of the target point b5 (i.e., obstacle point b5) is higher than that of the target point b4 (i.e., obstacle point b4) and the target point b6 (i.e., obstacle point b4), the longitudinal position coordinates of the missing trajectory point q1 determined at this time are the longitudinal position coordinates of the target point b5; the lateral position coordinates of the missing trajectory point q1 are the lateral safety distance.

[0074] This scheme determines the missing trajectory points by determining a dangerous point from multiple target points. There is no need to determine a corresponding missing trajectory point for each target point. While correcting the predicted driving trajectory to cross the obstacle boundary, the number of missing trajectory points added is reduced. That is, the original appearance of the predicted driving trajectory is maintained by minimizing the modification of the predicted driving trajectory as much as possible.

[0075] S304: adding the missing trajectory points to the predicted driving trajectory according to the position information of the missing trajectory points.

[0076] Specifically, this embodiment can find the position point corresponding to the position information in the middle of the target trajectory point pair corresponding to the missing trajectory point based on the position information of each missing trajectory point determined in S303, and add a trajectory point at the position point. After all the missing trajectory points are added, the predicted driving trajectory points can be corrected.

[0077] Optionally, after executing this step and adding the missing trajectory points to the predicted driving trajectory according to the position information of the missing trajectory points, this embodiment can also determine whether the target trajectory point pair still exists in the predicted driving trajectory after the missing trajectory points are added, in a manner similar to S302 above. If so, the target trajectory point at this time is determined, and a new missing trajectory point is determined again in a manner similar to S303-S304 and added to the predicted driving trajectory, until the target trajectory point pair no longer exists in the predicted driving trajectory after the missing trajectory points are added.

[0078] It should be noted that this embodiment mainly corrects the situation of crossing the obstacle boundary in the predicted driving trajectory. In addition, the present disclosure can also combine this embodiment with the above embodiments to correct both the situation of being too close to the obstacle boundary and the situation of crossing the obstacle boundary in the predicted driving trajectory. At this time, in order to ensure the accuracy of the correction result, the situation of being too close to the obstacle boundary in the predicted driving trajectory can be corrected first, and then the situation of crossing the obstacle boundary can be corrected.

[0079] The solution of the embodiment of the present disclosure obtains the position information of the obstacle points in the obstacle boundary of the section where the predicted driving trajectory of the driving object is located, and determines the target trajectory point pair and the target points included therein according to the position information of the obstacle points and the position information of the candidate trajectory points in the predicted driving trajectory. According to the position information of the target points and the lateral safety distance, the position information of the missing trajectory points is first determined, and then the missing trajectory points are added to the predicted driving trajectory. This solution provides a way to accurately determine the position of the missing trajectory points in combination with the obstacle boundary, solves the problem that the predicted driving trajectory crosses the obstacle boundary, and greatly improves the accuracy of the predicted driving trajectory.

[0080] Figure 4A It is a schematic diagram of a method for correcting a driving trajectory according to an embodiment of the present disclosure. Figure 4B It is a schematic diagram of the principle of screening missing trajectory points provided by an embodiment of the present disclosure. On the basis of the above embodiment, the embodiment of the present disclosure further explains in detail how to add missing trajectory points to the predicted driving trajectory according to the position information of the missing trajectory points, as Figure 4A-4B shown, the method for correcting the driving trajectory provided by this embodiment may include:

[0081] S401, obtain the obstacle boundary information of the section where the predicted driving trajectory of the driving object is located.

[0082] Among them, the obstacle boundary information at least includes the position information of the obstacle points in the obstacle boundary.

[0083] S402, determine the target trajectory point pair in the predicted driving trajectory and the target points included between the target trajectory point pairs according to the position information of the obstacle points in the obstacle boundary and the position information of the candidate trajectory points in the predicted driving trajectory.

[0084] Among them, the target points belong to the obstacle points.

[0085] S403, determine the position information of the missing trajectory points according to the position information of the target points and the lateral safety distance of the driving object when driving in the section.

[0086] S404. Determine whether the geometric relationship between the position information of the missing trajectory point and the position information of the target trajectory point pair satisfies the addition rule. If it satisfies, execute S405; if it does not satisfy, execute S406.

[0087] Specifically, one implementable manner of this embodiment is as follows: According to the position information of the missing trajectory point and the position information of the target trajectory point pair, determine whether the missing trajectory point is located between the connection line of the target trajectory point pair and the lane center line. If so, it satisfies the addition rule; if not, it does not satisfy the addition rule.

[0088] Exemplarily, as Figure 4B shown, p3 and p4 are a group of target trajectory point pairs, b4 - b6 are three target points located between this group of target trajectory point pairs, q1 and q2 are respectively the position information of two missing trajectory points determined by S403 based on these three target points. As Figure 4B shown, q1 is located on the left side of the connection line of p3 and p4 (i.e., between the lane center line and the connection line of p3 and p4), while q2 is located on the right side of the connection line of p3 and p4 (i.e., between the road boundary line and the connection line of p3 and p4). Then, at this time, it can be determined that the geometric relationship between the position information of the missing trajectory point q1 and the position information of the target trajectory point pair p3 and p4 satisfies the addition rule, while the geometric relationship between the position information of the missing trajectory point q2 and the position information of the target trajectory point pair p3 and p4 does not satisfy the addition rule.

[0089] Another implementable manner of this embodiment is as follows: According to the position information of the first trajectory point and the second trajectory point in the target trajectory point pair, determine the first slope value; according to the position information of the missing trajectory point and the position information of the first trajectory point, determine the second slope value; according to the first slope value and the second slope value, determine whether the geometric relationship between the position information of the missing trajectory point and the position information of the target trajectory point pair satisfies the addition rule; where the timestamp of the first trajectory point is earlier than the timestamp of the second trajectory point, that is, the driving object arrives at the first trajectory point first, and then arrives at the second trajectory point.

[0090] Specifically, if the position coordinates of the first trajectory point are (l1, s1), the position coordinates of the second trajectory point are (l2, s2); the position coordinates of the missing trajectory point are (l3, s3); then the first slope K1 = (l2 - l1) / (s2 - s1); the second slope K1 = (l3 - l1) / (s3 - s1); if the obstacle boundary is on the right side of the road section, then both the first slope and the second slope are less than 0; at this time, if the second slope value K2 is greater than the first slope value K1, it satisfies the addition rule; if the obstacle boundary is on the left side of the road section, then both the first slope and the second slope are greater than 0; at this time, if the second slope value K2 is less than the first slope value K1, it satisfies the addition rule.

[0091] Exemplarily, as Figure 4B known, Kq1p4 >K p3p4 Then, the geometric relationship between the position information of the missing trajectory point q1 and the position information of the target trajectory point pair (i.e., p3 and p4) satisfies the addition rule, K q2p4 <K p3p4 Then, the geometric relationship between the position information of the missing trajectory point q2 and the position information of the target trajectory point pair (i.e., p3 and p4) does not satisfy the addition rule.

[0092] In this embodiment, the slope relationship between the missing trajectory point and the target trajectory point pair is used to characterize the geometric relationship between the missing trajectory point and the target trajectory point pair, which improves the accuracy of the geometric relationship characterization, and further improves the judgment efficiency and accuracy of whether the addition rule is satisfied subsequently.

[0093] S405. If it is satisfied, then according to the position information of the missing trajectory point, add the missing trajectory point to the predicted driving trajectory.

[0094] S406. If it is not satisfied, then do not add the missing trajectory point to the predicted driving trajectory.

[0095] The solution of the embodiment of the present disclosure obtains the position information of the obstacle points in the obstacle boundary of the section where the predicted driving trajectory of the driving object is located, and determines the target trajectory point pair and the target points included therein according to the position information of the obstacle points and the position information of the candidate trajectory points in the predicted driving trajectory. According to the position information of the target points and the lateral safety distance, first determine the position information of the missing trajectory point, and only add the missing trajectory point to the predicted driving trajectory when the geometric relationship between the missing trajectory point and the target trajectory point pair satisfies the addition rule. This solution only adds the missing trajectory points whose geometric relationship satisfies the addition rule to the predicted driving trajectory, further ensuring the accuracy of adding the missing trajectory points.

[0096] Figure 5 is a flowchart of a method for correcting a driving trajectory provided according to an embodiment of the present disclosure. On the basis of the above embodiment, the embodiment of the present disclosure further explains in detail how to correct the predicted driving trajectory according to the position information of the obstacle points in the obstacle boundary, the position information of the candidate trajectory points in the predicted driving trajectory, and the lateral safety distance of the driving object when driving on the section, as Figure 5 shown, the method for correcting a driving trajectory provided in this embodiment may include:

[0097] S501. Obtain the obstacle boundary information of the section where the predicted driving trajectory of the driving object is located.

[0098] Among them, the obstacle boundary information at least includes the position information of the obstacle points in the obstacle boundary.

[0099] S502. Determine whether the predicted driving trajectory needs to be corrected according to the obstacle boundary type and the driving object type. If it is necessary, execute S503; if not, execute S504.

[0100] Optionally, in this embodiment, for different types of obstacle boundaries, combined with their obstacle characteristics, it can be preset whether to allow approaching driving and whether to allow crossing driving. If approaching driving is allowed, the type of driving object allowed to approach can be further limited. If crossing driving is possible, the maximum driving width allowed to cross can be further limited.

[0101] For example, if the obstacle boundary is a curb, pedestrians are allowed to approach, and vehicles are not allowed to approach; if the obstacle boundary is a cone, the maximum driving width allowed to cross is the distance between adjacent cones.

[0102] At this time, in this step, it can be determined according to the obstacle boundary type whether the obstacle boundary allows approaching driving. If not, execute subsequent S503 to correct the situation where the predicted driving trajectory is too close to the obstacle boundary. If it is allowed, further judge according to the driving object type whether the driving object is allowed to approach the obstacle boundary for driving. If not, execute subsequent S503 to correct the situation where the predicted driving trajectory is too close to the obstacle boundary; otherwise, execute S504 to end the correction of the predicted driving trajectory.

[0103] It can also be determined according to the obstacle boundary type whether the obstacle boundary allows crossing driving. If not, execute subsequent S503 to correct the situation where the predicted driving trajectory crosses the obstacle boundary; if it is allowed, then determine the driving width according to the driving object type. If the driving width is greater than the maximum driving width allowed to cross by the obstacle boundary, execute subsequent S503 to correct the situation where the predicted driving trajectory crosses the obstacle boundary; otherwise, execute S504 to end the correction of the predicted driving trajectory.

[0104] S503. If necessary, correct the predicted driving trajectory according to the position information of the obstacle points in the obstacle boundary, the position information of the candidate trajectory points in the predicted driving trajectory, and the lateral safety distance of the driving object during driving on the road section.

[0105] It should be noted that this embodiment can adopt any one of the correction methods introduced in the above embodiments, or a combination of multiple correction methods, to execute the correction operation of the predicted driving trajectory in this step, and no limitation is imposed on this.

[0106] S503. If not necessary, end the correction of the predicted driving trajectory.

[0107] Before correcting the predicted driving trajectory based on the position information of the obstacle points in the obtained obstacle boundary, the position information of the candidate trajectory points in the predicted driving trajectory, and the lateral safety distance of the driving object during driving on the road section, it is first determined whether this correction is required according to the type of the obstacle boundary and the type of the driving object. If it is required, subsequent trajectory correction operations are performed, avoiding incorrect corrections or unnecessary corrections, etc., and further ensuring the rationality and accuracy of the trajectory correction process.

[0108] Figure 6 FIG. 4 is a schematic structural diagram of a driving trajectory correction device provided by an embodiment of the present disclosure. The embodiment of the present disclosure is applicable to the situation of correcting a predicted driving trajectory. In particular, it is applicable to the situation of correcting a driving trajectory predicted by a deep learning model for a driving object. The device can be configured in an electronic device with a driving trajectory prediction or usage function, and is implemented by software and / or hardware. The device can implement the driving trajectory correction method of any embodiment of the present disclosure. As Figure 6 shown, the driving trajectory correction device 600 includes:

[0109] A boundary information acquisition module 601, configured to acquire obstacle boundary information of a road section where a predicted driving trajectory of a driving object is located; wherein, the obstacle boundary information at least includes the position information of obstacle points in the obstacle boundary;

[0110] A trajectory correction module 602, configured to correct the predicted driving trajectory according to the position information of the obstacle points in the obstacle boundary, the position information of the candidate trajectory points in the predicted driving trajectory, and the lateral safety distance of the driving object during driving on the road section.

[0111] The solution of the embodiment of the present disclosure acquires the position information of the obstacle points in the obstacle boundary of the road section where the predicted driving trajectory of the driving object is located, and corrects the predicted driving trajectory according to the position information of the obstacle points, the position information of the candidate trajectory points in the predicted driving trajectory, and the lateral safety distance of the driving object during driving on the road section. This solution can correct the error of the predicted driving trajectory, and during the process of correcting the predicted driving trajectory, it focuses on considering the obstacle boundary information of the road section, effectively reducing unreasonable problems such as the predicted trajectory being too close to the obstacle boundary or crossing the obstacle boundary, improving the accuracy of the predicted driving trajectory, and further improving the safety of autonomous driving.

[0112] Further, the trajectory correction module 602 includes:

[0113] An error trajectory point determination unit, configured to determine error trajectory points among the candidate trajectory points according to the position information of obstacle points in the obstacle boundary, the position information of candidate trajectory points in the predicted driving trajectory, and the lateral safety distance of the driving object when driving on the road section;

[0114] A position correction unit, configured to perform position correction on the error trajectory points in the predicted driving trajectory according to the lateral safety distance.

[0115] Further, the error trajectory point determination unit is specifically configured to:

[0116] Determine the lateral predicted distance between the candidate trajectory points and the obstacle boundary according to the position information of obstacle points in the obstacle boundary and the position information of candidate trajectory points in the predicted driving trajectory;

[0117] If the lateral predicted distance is less than the lateral safety distance of the driving object when driving on the road section, then use the candidate trajectory points as error trajectory points.

[0118] Further, the trajectory correction module 602 includes:

[0119] A target point determination unit, configured to determine target trajectory point pairs in the predicted driving trajectory and target points included between the target trajectory point pairs according to the position information of obstacle points in the obstacle boundary and the position information of candidate trajectory points in the predicted driving trajectory; wherein, the target points belong to the obstacle points;

[0120] A missing point determination unit, configured to determine the position information of missing trajectory points according to the position information of the target points and the lateral safety distance of the driving object when driving on the road section;

[0121] A missing point addition unit, configured to add missing trajectory points to the predicted driving trajectory according to the position information of the missing trajectory points.

[0122] Further, the missing point determination unit is specifically configured to:

[0123] Determine the risk coefficient of the target points according to the position information of the target points;

[0124] Determine dangerous points from the target points according to the risk coefficients of the target points;

[0125] Determine the position information of the missing trajectory points according to the position information of the dangerous points and the lateral safety distance of the driving object when driving on the road section.

[0126] Further, the missing point addition unit includes:

[0127] A rule judgment subunit, configured to determine whether the geometric relationship between the position information of the missing trajectory point and the position information of the target trajectory point pair satisfies an addition rule;

[0128] A missing point addition subunit, configured to, if satisfied, add the missing trajectory point to the predicted driving trajectory according to the position information of the missing trajectory point.

[0129] Further, the rule judgment subunit is specifically configured to:

[0130] Determine a first slope value according to the position information of the first trajectory point and the second trajectory point in the target trajectory point pair;

[0131] Determine a second slope value according to the position information of the missing trajectory point and the position information of the first trajectory point;

[0132] Determine whether the geometric relationship between the position information of the missing trajectory point and the position information of the target trajectory point pair satisfies the addition rule according to the first slope value and the second slope value;

[0133] Wherein, the timestamp of the first trajectory point is earlier than the timestamp of the second trajectory point.

[0134] Further, the obstacle boundary information further includes an obstacle boundary type;

[0135] The trajectory correction module 602 is further configured to:

[0136] Determine whether it is necessary to correct the predicted driving trajectory according to the obstacle boundary type and the driving object type;

[0137] If necessary, correct the predicted driving trajectory according to the position information of the obstacle points in the obstacle boundary, the position information of the candidate trajectory points in the predicted driving trajectory, and the lateral safety distance of the driving object when driving on the road section.

[0138] Further, the device 600 further includes:

[0139] A driving width determination module, configured to determine the driving width of the driving object according to the driving object type;

[0140] A safety distance determination module, configured to determine the lateral safety distance of the driving object when driving on the road section according to the driving width and the buffer distance.

[0141] Further, the device 600 further includes:

[0142] A buffer distance determination module, configured to determine a buffer distance according to the road section width of the road section, the obstacle boundary type in the obstacle boundary information, and the type of the driving object.

[0143] The above product can execute the method provided by any embodiment of the present disclosure, and has the corresponding functional modules and beneficial effects for executing the method.

[0144] In the technical solution of the present disclosure, the acquisition, storage, and application of the predicted driving trajectory, the obstacle boundary information of the road section, the type of the driving object, etc. all comply with the provisions of relevant laws and regulations and do not violate public order and good customs.

[0145] According to an embodiment of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium, and a computer program product.

[0146] Figure 7 FIG. shows a schematic block diagram of an exemplary electronic device 700 that can be used to implement the embodiments of the present disclosure. The electronic device is intended to represent various forms of digital computers, such as, a laptop computer, a desktop computer, a workbench, a personal digital assistant, a server, a blade server, a mainframe computer, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as, a personal digital assistant, a cellular phone, a smart phone, a wearable device, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely exemplary and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0147] As Figure 7 shown, the device 700 includes a computing unit 701, which can execute various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 702 or a computer program loaded from a storage unit 708 into a random access memory (RAM) 703. In the RAM 703, various programs and data required for the operation of the device 700 can also be stored. The computing unit 701, the ROM 702, and the RAM 703 are connected to each other through a bus 704. An input / output (I / O) interface 705 is also connected to the bus 704.

[0148] A plurality of components in the device 700 are connected to the I / O interface 705, including: an input unit 706, such as a keyboard, a mouse, etc.; an output unit 707, such as various types of displays, speakers, etc.; a storage unit 708, such as a magnetic disk, an optical disk, etc.; and a communication unit 709, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 709 allows the device 700 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0149] The computing unit 701 can be various general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 701 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 701 executes the various methods and processes described above, such as the method for correcting the driving trajectory. For example, in some embodiments, the method for correcting the driving trajectory can be implemented as a computer software program tangibly embodied in a machine-readable medium, such as the storage unit 708. In some embodiments, part or all of the computer program can be loaded and / or installed onto the device 700 via the ROM 702 and / or the communication unit 709. When the computer program is loaded into the RAM 703 and executed by the computing unit 701, one or more steps of the method for correcting the driving trajectory described above can be executed. Alternatively, in other embodiments, the computing unit 701 can be configured to execute the method for correcting the driving trajectory by any other suitable means (e.g., by means of firmware).

[0150] The various embodiments of the systems and techniques described above in this document can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGA), application-specific integrated circuits (ASIC), application-specific standard products (ASSP), system-on-a-chip systems (SOC), complex programmable logic devices (CPLD), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a dedicated or general-purpose programmable processor, and can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit the data and instructions to the storage system, the at least one input device, and the at least one output device.

[0151] The program code for implementing the methods of the present disclosure can be written in any combination of one or more programming languages. These program codes can be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when the program codes are executed by the processor or controller, the functions / operations specified in the flowcharts and / or block diagrams are implemented. The program code can be executed entirely on the machine, partially on the machine, as a stand-alone software package partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0152] In the context of this disclosure, a machine-readable medium can be a tangible medium that can contain or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. The machine-readable medium can include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of the machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0153] To provide for interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the computer. Other kinds of devices can also be used to provide for interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic, speech, or tactile input).

[0154] The systems and techniques described herein can be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer having a graphical user interface or a web browser through which the user can interact with an implementation of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local area network (LAN), wide area network (WAN), blockchain network, and the Internet.

[0155] A computer system may include a client and a server. The client and the server are generally far from each other and usually interact via a communication network. The relationship between the client and the server is generated by computer programs running on respective computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or a cloud host, which is a host product in the cloud computing service system, solving the defects of difficult management and weak business scalability existing in traditional physical hosts and VPS services. The server may also be a server of a distributed system or a server combined with blockchain.

[0156] Artificial intelligence is a discipline that studies how to make a computer simulate certain thinking processes and intelligent behaviors of humans (such as learning, reasoning, thinking, planning, etc.), and it has technologies at both the hardware level and the software level. Artificial intelligence hardware technologies generally include technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, and big data processing; artificial intelligence software technologies mainly include several major directions such as computer vision technology, speech recognition technology, natural language processing technology, machine learning / deep learning technology, big data processing technology, and knowledge graph technology.

[0157] Cloud computing refers to a technical system that accesses an elastic and scalable shared physical or virtual resource pool through a network. The resources can include servers, operating systems, networks, software, applications, and storage devices, etc., and the resources can be deployed and managed in a demand-based and self-service manner. Through cloud computing technology, it can provide efficient and powerful data processing capabilities for the application and model training of technologies such as artificial intelligence and blockchain.

[0158] In addition, an embodiment of the present invention also provides an autonomous vehicle, including a vehicle body, and the electronic device provided in the above embodiment of the present invention is arranged on the vehicle body.

[0159] It should be understood that various forms of the processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this disclosure can be executed in parallel, sequentially, or in a different order, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved, and no limitations are imposed herein.

[0160] The above specific embodiments do not constitute a limitation on the protection scope of this disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure shall be included within the protection scope of this disclosure.

Claims

1. A method for correcting a driving trajectory, comprising: obtaining obstacle boundary information of a section where a predicted driving trajectory of a driving object is located; wherein, the obstacle boundary information at least includes position information of obstacle points in the obstacle boundary; correcting the predicted driving trajectory according to the position information of the obstacle points in the obstacle boundary, the position information of candidate trajectory points in the predicted driving trajectory, and the lateral safety distance of the driving object when driving in the section; The correcting the predicted driving trajectory according to the position information of the obstacle points in the obstacle boundary, the position information of candidate trajectory points in the predicted driving trajectory, and the lateral safety distance of the driving object when driving in the section includes: determining target trajectory point pairs in the predicted driving trajectory and target points included between the target trajectory point pairs according to the position information of the obstacle points in the obstacle boundary and the position information of candidate trajectory points in the predicted driving trajectory; wherein, the target points belong to the obstacle points; determining the position information of missing trajectory points according to the position information of the target points and the lateral safety distance of the driving object when driving in the section; wherein, the longitudinal position coordinate of the missing trajectory point is the longitudinal position coordinate of the target point; the lateral position coordinate of the missing trajectory point is the lateral safety distance; adding missing trajectory points to the predicted driving trajectory according to the position information of the missing trajectory points.

2. The method according to claim 1, wherein, the correcting the predicted driving trajectory according to the position information of the obstacle points in the obstacle boundary, the position information of candidate trajectory points in the predicted driving trajectory, and the lateral safety distance of the driving object when driving in the section includes: determining error trajectory points among the candidate trajectory points according to the position information of the obstacle points in the obstacle boundary, the position information of candidate trajectory points in the predicted driving trajectory, and the lateral safety distance of the driving object when driving in the section; correcting the positions of the error trajectory points in the predicted driving trajectory according to the lateral safety distance.

3. The method according to claim 2, wherein, the determining error trajectory points among the candidate trajectory points according to the position information of the obstacle points in the obstacle boundary, the position information of candidate trajectory points in the predicted driving trajectory, and the lateral safety distance of the driving object when driving in the section includes: determining the lateral predicted distance between the candidate trajectory points and the obstacle boundary according to the position information of the obstacle points in the obstacle boundary and the position information of candidate trajectory points in the predicted driving trajectory; if the lateral predicted distance is less than the lateral safety distance of the driving object when driving in the section, then taking the candidate trajectory point as an error trajectory point.

4. The method according to claim 1, wherein, the determining the position information of missing trajectory points according to the position information of the target points and the lateral safety distance of the driving object when driving in the section includes: determining the risk coefficient of the target point according to the position information of the target point; determining dangerous points from the target points according to the risk coefficient of the target point; Determine the position information of the missing trajectory point according to the position information of the dangerous point and the lateral safety distance of the driving object when driving on the road section.

5. The method according to claim 1, wherein, Adding a missing trajectory point to the predicted driving trajectory according to the position information of the missing trajectory point includes: Determine whether the geometric relationship between the position information of the missing trajectory point and the position information of the target trajectory point pair satisfies the addition rule; If it is satisfied, add the missing trajectory point to the predicted driving trajectory according to the position information of the missing trajectory point.

6. The method according to claim 5, wherein, Determining whether the geometric relationship between the position information of the missing trajectory point and the position information of the target trajectory point pair satisfies the addition rule includes: Determine the first slope value according to the position information of the first trajectory point and the second trajectory point in the target trajectory point pair; Determine the second slope value according to the position information of the missing trajectory point and the position information of the first trajectory point; Determine whether the geometric relationship between the position information of the missing trajectory point and the position information of the target trajectory point pair satisfies the addition rule according to the first slope value and the second slope value; wherein, the timestamp of the first trajectory point is earlier than the timestamp of the second trajectory point.

7. The method according to any one of claims 1-6, wherein, The obstacle boundary information further includes an obstacle boundary type; Correspondingly, the correcting the predicted driving trajectory according to the position information of the obstacle points in the obstacle boundary, the position information of the candidate trajectory points in the predicted driving trajectory, and the lateral safety distance of the driving object when driving on the road section further includes: Determine whether it is necessary to correct the predicted driving trajectory according to the obstacle boundary type and the driving object type; If necessary, correct the predicted driving trajectory according to the position information of the obstacle points in the obstacle boundary, the position information of the candidate trajectory points in the predicted driving trajectory, and the lateral safety distance of the driving object when driving on the road section.

8. The method according to any one of claims 1-7, further includes: Determine the driving width of the driving object according to the driving object type; Determine the lateral safety distance of the driving object when driving on the road section according to the driving width and the buffer distance.

9. The method according to claim 8, further includes: Determine the buffer distance according to the road section width of the road section, the obstacle boundary type in the obstacle boundary information, and the driving object type.

10. A device for correcting a driving trajectory, includes: A boundary information acquisition module, configured to acquire obstacle boundary information of a road section where a predicted driving trajectory of a driving object is located; wherein, the obstacle boundary information at least includes the position information of obstacle points in the obstacle boundary; A trajectory correction module, configured to correct the predicted driving trajectory according to the position information of the obstacle points in the obstacle boundary, the position information of the candidate trajectory points in the predicted driving trajectory, and the lateral safety distance of the driving object when driving on the road section; Among them, the trajectory correction module includes a target point determination unit, a missing point determination unit, and a missing point addition unit; The target point determination unit is configured to determine a target trajectory point pair in the predicted driving trajectory and target points included between the target trajectory point pairs according to the position information of obstacle points in the obstacle boundary and the position information of candidate trajectory points in the predicted driving trajectory; among them, the target points belong to the obstacle points; The missing point determination unit is configured to determine the position information of missing trajectory points according to the position information of the target points and the lateral safety distance of the driving object when driving on the section; among them, the longitudinal position coordinate of the missing trajectory point is the longitudinal position coordinate of the target point; the lateral position coordinate of the missing trajectory point is the lateral safety distance; The missing point addition unit is configured to add missing trajectory points to the predicted driving trajectory according to the position information of the missing trajectory points.

11. The device according to claim 10, wherein, the trajectory correction module includes: an error trajectory point determination unit configured to determine error trajectory points among the candidate trajectory points according to the position information of obstacle points in the obstacle boundary, the position information of candidate trajectory points in the predicted driving trajectory, and the lateral safety distance of the driving object when driving on the section; a position correction unit configured to correct the positions of the error trajectory points in the predicted driving trajectory according to the lateral safety distance.

12. The device according to claim 11, wherein, the error trajectory point determination unit is specifically configured to: determine the lateral predicted distance between the candidate trajectory points and the obstacle boundary according to the position information of obstacle points in the obstacle boundary and the position information of candidate trajectory points in the predicted driving trajectory; if the lateral predicted distance is less than the lateral safety distance of the driving object when driving on the section, then use the candidate trajectory points as error trajectory points.

13. The device according to claim 10, wherein, the missing point determination unit is specifically configured to: determine the risk coefficient of the target points according to the position information of the target points; determine dangerous points from the target points according to the risk coefficient of the target points; determine the position information of the missing trajectory points according to the position information of the dangerous points and the lateral safety distance of the driving object when driving on the section.

14. The device according to claim 10, wherein, the missing point addition unit includes: a rule judgment sub-unit configured to determine whether the geometric relationship between the position information of the missing trajectory points and the position information of the target trajectory point pair satisfies the addition rule; a missing point addition sub-unit configured to, if satisfied, add the missing trajectory points to the predicted driving trajectory according to the position information of the missing trajectory points.

15. The device according to claim 14, wherein, the rule judgment sub-unit is specifically configured to: determine a first slope value according to the position information of the first trajectory point and the second trajectory point in the target trajectory point pair; determine a second slope value according to the position information of the missing trajectory point and the position information of the first trajectory point; Determine whether the geometric relationship between the position information of the missing trajectory point and the position information of the target trajectory point pair satisfies the addition rule according to the first slope value and the second slope value; Wherein, the timestamp of the first trajectory point is earlier than the timestamp of the second trajectory point.

16. The apparatus according to any one of claims 10-15, Wherein, The obstacle boundary information further includes an obstacle boundary type; The trajectory correction module is further configured to: Determine whether it is necessary to correct the predicted driving trajectory according to the obstacle boundary type and the driving object type; If necessary, correct the predicted driving trajectory according to the position information of the obstacle points in the obstacle boundary, the position information of the candidate trajectory points in the predicted driving trajectory, and the lateral safety distance of the driving object when driving on the road section.

17. The apparatus according to any one of claims 10-16, further Comprises: A driving width determination module, configured to determine the driving width of the driving object according to the driving object type; A safety distance determination module, configured to determine the lateral safety distance of the driving object when driving on the road section according to the driving width and the buffer distance.

18. The apparatus according to claim 17, further Comprises: A buffer distance determination module, configured to determine a buffer distance according to the road section width of the road section, the obstacle boundary type in the obstacle boundary information, and the driving object type.

19. An electronic device, Comprises: At least one processor; And A memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the driving trajectory correction method according to any one of claims 1-9.

20. A non-transitory computer-readable storage medium storing computer instructions, Wherein, The computer instructions are used to cause the computer to execute the driving trajectory correction method according to any one of claims 1-9.

21. A computer program product, comprising a computer program which, when executed by a processor, implements the driving trajectory correction method according to any one of claims 1-9.

22. An autonomous vehicle, comprising the electronic device according to claim 19.

Images