Track smoothing method, vehicle, equipment and program product
By screening and dividing the trajectory of other vehicles, the trajectory discontinuity caused by noise information is solved, and the accuracy of driving decisions of autonomous vehicles is improved.
Patent Information
- Application Number
- CN202411545973.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-10-31
- Publication Date
- 2024-12-06
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
When an autonomous vehicle is processing the trajectory of other vehicles, the noise information causes the trajectory to be discontinuous, affecting the accuracy of driving decisions.
By filtering out continuous trajectories whose spacing between adjacent trajectories points is less than the set distance and disconnected trajectories whose spacing is greater than or equal to the set distance, dividing and smoothing processing are performed to generate a smooth target driving trajectory.
It improves the continuity and accuracy of the trajectory, reduces the impact of noise, and enhances the accuracy and rationality of bicycle decisions.
Smart Images

Figure CN119078886A_ABST
Abstract
Description
Technical Field
[0001] The present disclosure relates to the field of autonomous driving technology, and in particular to a trajectory smoothing method, a vehicle, a device, and a program product. Background Art
[0002] In the field of autonomous driving technology, autonomous vehicles need to use the collected road environment information around the vehicle and the vehicle trajectories of other vehicles to plan their own driving. However, the vehicle trajectories of other vehicles collected by the vehicle may contain certain noise, which affects the smoothness of the vehicle trajectory. If the noisy trajectories of other vehicles are directly processed, the accuracy of the driving decision planned by the vehicle will be affected. Therefore, how to smooth the trajectories of other vehicles to eliminate the impact of noise in the trajectories on driving decisions has become a technical issue worthy of attention. Summary of the invention
[0003] The embodiments of the present disclosure at least provide a trajectory smoothing method, a vehicle, a device, and a program product.
[0004] In a first aspect, an embodiment of the present disclosure provides a trajectory smoothing method, comprising: Acquire driving track information of each other vehicle detected by the self-vehicle; the driving track information includes first coordinate information of multiple track points in the historical driving track of the other vehicle; According to the first coordinate information of the plurality of trajectory points, a first driving trajectory and a second driving trajectory are selected from each of the historical driving trajectories; the distance between each adjacent trajectory point in the first driving trajectory is less than a set distance, and the distance between at least two adjacent trajectory points in the second driving trajectory is greater than or equal to the set distance; Using at least two adjacent track points in the second driving track whose distance is greater than or equal to a set distance, the second driving track is divided into at least two third driving tracks, and the first driving track and the third driving track are used as target driving tracks; The target driving trajectory is smoothed to obtain a smoothed driving trajectory.
[0005] In a possible implementation manner, the step of smoothing the target driving trajectory to obtain a smoothed driving trajectory includes: A plurality of control points are selected from the target driving trajectory, and a weight of each control point and a node vector composed of target nodes corresponding to the plurality of control points are determined according to a curve smoothing order and first coordinate information of the control point; each target node is determined according to at least one control point, and the node vector is used to control segments in the trajectory and the shape of each segment; A smoothed driving trajectory corresponding to the target driving trajectory is generated according to the node vector and the weight of the control point.
[0006] In a possible implementation manner, the step of selecting a plurality of control points from the target driving trajectory includes: According to the first coordinate information of each target trajectory point in the target driving trajectory, the previous trajectory point among the adjacent trajectory points whose interval is less than a first distance is removed from the target driving trajectory in sequence; the first distance is less than the set distance; According to the track length of the target driving track, a plurality of control points are screened out from the reserved track points in the target driving track.
[0007] In a possible implementation, according to the trajectory length of the target driving trajectory, a plurality of control points are screened out from the reserved trajectory points in the target driving trajectory, including: When the trajectory length is greater than or equal to the first length, selecting a plurality of control points from the reserved trajectory points in the target driving trajectory according to the first interval value; In the case where the trajectory length is less than or equal to the second length, all the reserved trajectory points in the target driving trajectory are used as the control points; the first length is greater than the second length; When the trajectory length is between the second length and the first length, a plurality of control points are screened out from the retained trajectory points in the target driving trajectory according to a second interval value; the second interval value is smaller than the first interval value.
[0008] In a possible implementation manner, after selecting a plurality of control points from the target driving trajectory, the method further includes: When the number of the plurality of control points is less than or equal to the curve smoothing order, all the reserved track points in the target driving track are used as the control points.
[0009] In a possible implementation, generating a smoothed driving trajectory corresponding to the target driving trajectory according to the node vector and the weight of the control point includes: According to the node vector, at least two new target nodes are inserted at equal intervals between every two target nodes, and the node vector is updated using each inserted target node; Drawing a non-uniform rational B-spline curve according to the weight of each control point and the updated node vector; Converting the first coordinate information of each curve point in the non-uniform rational B-spline curve into the second coordinate information in the longitude and latitude coordinate system; A smoothed driving trajectory corresponding to the target driving trajectory is generated according to the second coordinate information.
[0010] In a possible implementation manner, after generating a smoothed driving trajectory corresponding to the target driving trajectory according to the node vector and the weight of the control point, the method further includes: For any target curve point except the first curve point and the last curve point in the smoothed driving trajectory, determine the curvature of the target curve point according to the second coordinate information of the curve points adjacent to the target curve point in front and behind in the longitude and latitude coordinate system; Determining the curvature difference between every two adjacent target curve points according to the curvature of the target curve point; The curvature differences are sorted in order from small to large, and the curvature difference at the target percentile in the sorting order is used as the smoothness evaluation index of the smoothed driving trajectory.
[0011] In a possible implementation manner, taking the first driving trajectory and the third driving trajectory as target driving trajectories includes: Trajectories whose trajectory lengths are less than the second distance in the first driving trajectory and the third driving trajectory are eliminated, and the remaining trajectory is used as the target driving trajectory.
[0012] In a possible implementation manner, the step of obtaining the driving trajectory information of each other vehicle detected by the vehicle includes: Acquire initial trajectory information of each other vehicle detected by the self-vehicle, wherein the initial trajectory information includes third coordinate information of multiple trajectory points in the self-vehicle coordinate system; According to the conversion relationship between the vehicle coordinate system and the northeast sky coordinate system, the third coordinate information of the plurality of trajectory points is converted into the first coordinate information; The first coordinate information of multiple trajectory points is used as the driving trajectory information.
[0013] In a second aspect, an embodiment of the present disclosure further provides a vehicle, including a controller, wherein the controller includes: a memory for storing machine-readable instructions; and A processor is used to call and execute the machine-readable instructions stored in the memory to implement the above trajectory smoothing method, or steps in any possible implementation of the above trajectory smoothing method.
[0014] In a third aspect, an optional implementation of the present disclosure further provides a computer device, a processor, and a memory, wherein the memory stores machine-readable instructions executable by the processor, and the processor is used to execute the machine-readable instructions stored in the memory, and when the machine-readable instructions are executed by the processor, the steps of the above-mentioned first aspect, or any possible implementation of the first aspect are performed.
[0015] In a fourth aspect, an optional implementation of the present disclosure further provides a computer program product, which, when executed, executes the steps of the above-mentioned first aspect, or any possible implementation of the first aspect.
[0016] For a description of the effects of the above-mentioned vehicle, computer device, and computer program product, please refer to the description of the above-mentioned trajectory smoothing method, which will not be repeated here.
[0017] The trajectory smoothing method, vehicle, device and program product provided by the embodiments of the present disclosure can accurately screen out the continuous first driving trajectory with the distance between adjacent trajectory points less than the set distance and the disconnected second driving trajectory with the distance between adjacent trajectory points greater than or equal to the set distance from the historical driving trajectory of the other vehicle track by using the first coordinate information of the trajectory point. Then, by segmenting the disconnected second driving trajectory and smoothing each segmented trajectory, the problem of smoothing noise generated at the disconnected position when directly smoothing the disconnected trajectory can be avoided, which leads to an unreasonable smoothed trajectory, thereby improving the continuity of each smoothed trajectory output. Finally, the decision-making planning of the self-vehicle can be performed using the smoothed driving trajectory of the other vehicle, which can improve the accuracy and rationality of the decision.
[0018] In order to make the above-mentioned objectives, features and advantages of the present disclosure more obvious and easy to understand, preferred embodiments are specifically cited below and described in detail with reference to the attached drawings. BRIEF DESCRIPTION OF THE DRAWINGS
[0019] In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the following is a brief introduction to the drawings required for use in the embodiments. The drawings herein are incorporated into the specification and constitute a part of the specification. These drawings illustrate embodiments consistent with the present disclosure and are used together with the specification to illustrate the technical solutions of the present disclosure. It should be understood that the following drawings only illustrate certain embodiments of the present disclosure and should not be regarded as limiting the scope. For ordinary technicians in this field, other relevant drawings can also be obtained based on these drawings without creative work.
[0020] Figure 1 A flow chart of a trajectory smoothing method provided by an embodiment of the present disclosure is shown; Figure 2 A schematic diagram showing a comparison of historical driving trajectories before and after smoothing provided by an embodiment of the present disclosure is shown; Figure 3 A schematic diagram showing a specific implementation process of a trajectory smoothing method provided by an embodiment of the present disclosure is shown; Figure 4 A schematic diagram of the structure of a computer device provided by an embodiment of the present disclosure is shown. DETAILED DESCRIPTION
[0021] In order to make the purpose, technical scheme and advantages of the embodiments of the present disclosure clearer, the technical scheme in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are only part of the embodiments of the present disclosure, rather than all of the embodiments. The components of the embodiments of the present disclosure generally described and shown here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present disclosure is not intended to limit the scope of the present disclosure claimed for protection, but merely represents the selected embodiments of the present disclosure. Based on the embodiments of the present disclosure, all other embodiments obtained by those skilled in the art without making creative work belong to the scope of protection of the present disclosure.
[0022] In addition, the terms "first", "second", etc. in the specification and claims of the embodiments of the present disclosure and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. The "plurality or several" mentioned in this article refers to two or more. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the objects associated before and after are in an "or" relationship.
[0023] Through research, it is found that conventional trajectory smoothing methods include Kalman filtering, B-spline curve fitting, least squares curve fitting, etc. Although these methods can achieve smooth processing of trajectories, they cannot handle the noise information in the trajectory of other vehicles (for example, the trajectory is too short, the trajectory is disconnected, etc.), resulting in low accuracy and rationality of the smoothed curve. The accuracy of the smoothed trajectory plays a key role in improving the accuracy of motion planning and prediction. Therefore, a trajectory smoothing method that can effectively remove noise information is needed to output a complete, smooth, and continuous trajectory of other vehicles.
[0024] Based on the above research, the present disclosure provides a trajectory smoothing method, vehicle, device and program product. By using the first coordinate information of the trajectory point, the continuous first driving trajectory with the distance between adjacent trajectory points less than the set distance and the disconnected second driving trajectory with the distance between adjacent trajectory points greater than or equal to the set distance can be accurately screened from the historical driving trajectory of the other vehicle. Then, by segmenting the disconnected second driving trajectory and smoothing each segmented trajectory, the problem of smoothing noise generated at the disconnected position when directly smoothing the disconnected trajectory can be avoided, which leads to an unreasonable smoothed trajectory, and improves the continuity of each smoothed trajectory output. Finally, the decision-making planning of the self-vehicle can be performed using the smoothed driving trajectory of the other vehicle, which can improve the accuracy and rationality of the decision.
[0025] The defects existing in the above solutions are the results obtained by the inventor after practice and careful research. Therefore, the discovery process of the above problems and the solutions proposed by the present disclosure for the above problems below should be the contributions made by the inventor to the present disclosure during the disclosure process.
[0026] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, further definition and explanation thereof is not required in subsequent drawings.
[0027] It is understandable that before using the technical solutions disclosed in the embodiments of the present disclosure, the types, scope of use, usage scenarios, etc. of the personal information involved in the present disclosure should be informed to the user and the user's authorization should be obtained in an appropriate manner in accordance with relevant laws and regulations.
[0028] To facilitate understanding of this embodiment, a trajectory smoothing method disclosed in an embodiment of the present disclosure is first introduced in detail. The execution subject of the trajectory smoothing method provided in the embodiment of the present disclosure is generally a terminal device or other processing device with a certain computing capability, wherein the terminal device may be a user equipment (UE), a mobile device, a user terminal, a terminal, a personal digital assistant (PDA), a handheld device, a computer device, etc.; in some possible implementations, the trajectory smoothing method may be implemented by a processor calling a computer-readable instruction stored in a memory.
[0029] The trajectory smoothing method provided by the embodiment of the present disclosure is described below by taking a computer device as an example of an execution subject.
[0030] like Figure 1 As shown, it is a flow chart of a trajectory smoothing method provided by an embodiment of the present disclosure, which may include the following steps: S101: Acquire driving trajectory information of each other vehicle detected by the own vehicle; the driving trajectory information includes first coordinate information of multiple trajectory points in the historical driving trajectory of the other vehicle.
[0031] In a specific implementation, the self-vehicle may be a vehicle with an automatic driving function. The other vehicles may be various vehicles located around the self-vehicle detected by the self-vehicle during the driving process. The driving trajectory information may be trajectory information corresponding to the historical driving trajectory of the other vehicle, and the driving trajectory information of the other vehicle is also detected by the self-vehicle.
[0032] The driving trajectory information includes the coordinate information of multiple trajectory points in the historical driving trajectory of the other vehicle in the past period of time, and the coordinate information here is defined as the first coordinate information. The first coordinate information can be the coordinate value corresponding to the trajectory point in the east-north-sky coordinate system (ie, the ENU coordinate system). When the computer device distinguishes the historical driving trajectories of each other vehicle, the identifier of the historical driving trajectory of each other vehicle can be used, and the trajectory identifier of each other vehicle can use the unique identifier of the other vehicle. That is, the identifier of the trajectory of the same other vehicle is the same, and the identifier of the trajectory of different other vehicles is different. Therefore, the driving trajectory information of each other vehicle can also include the identifier of the historical driving trajectory of the other vehicle.
[0033] In a specific implementation, during the driving process of the vehicle, the vehicle can detect at least one other vehicle (such as other vehicles in front of the vehicle) that has an impact on the driving of the vehicle on the road and determine the identity of each other vehicle. For each other vehicle, the vehicle continuously collects each track point in the historical driving track of the other vehicle, and then determines the driving track information of the other vehicle based on the first coordinate information of each track point in the ENU coordinate, the order of each track point and the identity of the other vehicle.
[0034] S102: Filter out a first driving trajectory and a second driving trajectory from each historical driving trajectory according to first coordinate information of the plurality of trajectory points; the distance between each adjacent trajectory point in the first driving trajectory is less than a set distance, and the distance between at least two adjacent trajectory points in the second driving trajectory is greater than or equal to the set distance.
[0035] Here, the first driving trajectory is a historical driving trajectory in which the distance between any two adjacent trajectory points is less than the set distance. The second driving trajectory is a historical driving trajectory in which the distance between at least two adjacent trajectory points is not less than the set distance. The reason why it is limited to at least two here is that if the distance between only two adjacent trajectory points in the historical driving trajectory is not less than the set distance, the historical driving trajectory will be used as the second driving trajectory; for example, if the distance between only the third and fourth trajectory points in the historical driving trajectory is not less than the set distance, the historical driving trajectory will be used as the second driving trajectory. If the distances between multiple consecutive adjacent trajectory points in the historical driving trajectory are not less than the set distance, the historical driving trajectory will also be used as the second driving trajectory; for example, if the distance between each two adjacent trajectory points from the 1st trajectory point to the nth (n is greater than or equal to 3) trajectory point in the historical driving trajectory is not less than the set distance, the historical driving trajectory will be used as the second driving trajectory. If in the historical driving trajectory, there are multiple groups of adjacent track points whose distances are not less than the set distance, the historical driving trajectory will be used as the second driving trajectory; for example, in the historical driving trajectory, the distance between the 3rd track point and the 4th track point is not less than the set distance, the distance between the 8th track point and the 9th track point is not less than the set distance, and the distance between the 15th track point and the 16th track point is not less than the set distance. At this time, there are 3 groups of adjacent track points whose distances are less than the set distance, then the historical driving trajectory will also be used as the second driving trajectory.
[0036] The set distance can be set based on experience, and the embodiment of the present disclosure does not specifically limit it. For example, the set distance can be 5 meters (m), 10m, 20m, etc.
[0037] It should be noted that the reason why the distance between adjacent trajectory points is not less than the set distance is that there is a problem when the self-vehicle collects trajectory points, resulting in a gap; another reason is that during the driving process of the self-vehicle and the other car, for the other car, there may be situations such as "the other car is overtaken by the self-vehicle", "the other car surpasses the self-vehicle", "a new vehicle appears between the self-vehicle and the other car", etc., and the occurrence of these situations will cause the historical driving trajectory of the same other car collected by the self-vehicle to be disconnected in the middle. If the historical driving trajectory of the other car is less disconnected, for example, only one or two trajectory points are missing, the missing trajectory points have little effect on the smoothing of the trajectory and can be ignored. However, if the historical driving trajectory of the other car is more disconnected, for example, a long section is missing in the trajectory, the missing content will have a greater impact on the smoothing of the trajectory. If it is directly ignored, the smoothed curve will not reflect the real driving trajectory of the other car, which will affect the self-vehicle to make a correct decision. Therefore, before smoothing the historical driving trajectory of the other car, it is necessary to detect and divide the historical driving trajectory to determine whether there is a historical driving trajectory with a large interruption in the middle. If it occurs, the following S103 is used for processing.
[0038] In specific implementation, each historical driving trajectory can be distinguished by using the identifier of each historical driving trajectory. For each historical driving trajectory, the distance between each two adjacent trajectory points can be determined according to the first coordinate information of each trajectory point in the historical driving trajectory and the order of each trajectory point in the historical driving trajectory. Then, it is determined whether the distance between each two adjacent trajectory points in the historical driving trajectory is less than the set distance. If so, the historical driving trajectory is used as the first driving trajectory; if not, the historical driving trajectory is used as the second driving trajectory.
[0039] It is understandable that when the first driving trajectory and the second driving trajectory are divided according to the first coordinate information of the trajectory point, it is also possible that each historical driving trajectory is the first driving trajectory or each historical driving trajectory is the second driving trajectory. In the case where each historical driving trajectory is the first driving trajectory, each first driving trajectory can be directly used as the target driving trajectory. In the case where each historical driving trajectory is the second driving trajectory, each second driving trajectory can be divided using the following S103, and each trajectory obtained by the division is used as the target driving trajectory.
[0040] In this way, by splitting the historical driving trajectories whose distances between adjacent trajectory points are not less than the set distance, the emergence of large-span trajectories can be avoided, and the influence of large-span trajectories on the driving decision of the vehicle can be avoided.
[0041] S103: using at least two adjacent trajectory points in the second driving trajectory whose distance is greater than or equal to the set distance, dividing the second driving trajectory into at least two third driving trajectories, and taking the first driving trajectory and the third driving trajectory as target driving trajectories.
[0042] Here, the third driving track is a track cut out from the second driving track. A second driving track can be cut into at least two third driving tracks.
[0043] In a specific implementation, for each second driving trajectory, each group of adjacent trajectory points can be screened out from the second driving trajectory, wherein each group of adjacent trajectory points includes two target trajectory points, and the distance between the two adjacent trajectory points is greater than or equal to the set distance. Then, in the case where the adjacent trajectory points include a group, the first trajectory point in the group of adjacent trajectory points can be used as a segmentation point to segment the second driving trajectory, thereby obtaining a trajectory before the first trajectory point, and the trajectory is used as a third driving trajectory. At the same time, the trajectory after the second trajectory point in the group of adjacent trajectory points is used as another third driving trajectory.
[0044] In the case where the adjacent trajectory points include multiple groups, the second driving trajectory can be divided in sequence according to the order of each group of adjacent trajectory points in the second driving trajectory. Specifically, the second driving trajectory can be divided by using the previous trajectory point in the first group of adjacent trajectory points as a dividing point to obtain a third driving trajectory located before the first trajectory point, and then the trajectory after the next trajectory point in the first group of adjacent trajectory points to the previous trajectory point in the next group of adjacent trajectory points is used as a third driving trajectory; and the trajectory after the next trajectory point in the next group of adjacent trajectory points to the previous trajectory point in the new next group of adjacent trajectory points is used as a third driving trajectory, and so on and so forth, that is, the second driving trajectory can be divided into multiple third driving trajectories. That is, for the second driving trajectory, the second driving trajectory is divided at the position where the distance between the adjacent trajectory points is greater than or equal to the set distance to obtain each third driving trajectory.
[0045] Wherein, for each second driving track, after the second driving track is divided into different third driving tracks, the identifier of each third driving track can be determined by using the identifier of the second driving track and the preset identifier offset, and the identifiers of each third driving track are different. For example, the identifier of the second driving track is id1, the preset identifier offset is id_offset, the identifier of the first third driving track divided from the second driving track can be id1, the identifier of the second third driving track divided from the second driving track can be id1+id_offset, the identifier of the third third driving track divided from the second driving track can be id1+id_offset+id_offset, and so on, to obtain the identifier of each third driving track.
[0046] After the third driving trajectories corresponding to the second driving trajectories are divided, each of the divided third driving trajectories and each of the first driving trajectories can be used as a target driving trajectory.
[0047] S104: Smoothing the target driving trajectory to obtain a smoothed driving trajectory.
[0048] Exemplarily, for each target driving trajectory, any trajectory smoothing method in the prior art may be used to smooth the target driving trajectory to obtain each smoothed driving trajectory.
[0049] In this way, by using the first coordinate information of the trajectory point, the continuous first driving trajectory with the distance between adjacent trajectory points less than the set distance and the disconnected second driving trajectory with the distance between adjacent trajectory points greater than or equal to the set distance can be accurately screened from the historical driving trajectory of the other vehicle. Then, by segmenting the disconnected second driving trajectory and smoothing each segmented trajectory, the problem of smoothing noise generated at the disconnected position when directly smoothing the disconnected trajectory can be avoided, which leads to unreasonable smoothed trajectory, and improves the continuity of each smoothed trajectory output. Finally, the decision-making planning of the self-vehicle can be carried out using the smoothed driving trajectory of the other vehicle, which can improve the accuracy and rationality of the decision.
[0050] In one embodiment, the above S104 can be implemented according to the following steps: S104-1: Filter out multiple control points from the target driving trajectory, and determine the weight of each control point and the node vector composed of the target nodes corresponding to the multiple control points according to the curve smoothing order and the first coordinate information of the control point; each target node is determined based on at least one control point, and the node vector is used to control the segments in the trajectory and the shape of each segment.
[0051] Here, the embodiment of the present disclosure uses a non-uniform rational B-spline (NURBS) curve for trajectory smoothing, because during smoothing, it is necessary to first select control points for drawing the curve, and then determine the relevant information of each target node on the curve based on the control points. The curve smoothing order is the curve order corresponding to the NURBS curve method used in the embodiment of the present disclosure. The order can be set based on experience, and the embodiment of the present disclosure does not specifically limit it. For example, the curve smoothing order can be 3, that is, a 3rd-order NURBS curve is used for trajectory smoothing.
[0052] The control point is a point selected from each track point included in the target driving track, which is used to determine the target node. The target node is a point used to determine each key position point of the curve when drawing the NURBS curve. The number of target nodes is determined according to the curve smoothing order and the number of control points. Generally, the number of target nodes is equal to the number of control points plus the curve smoothing order minus 1. In the NURBS curve, the parameterized point K corresponding to the control point is called a node, that is, a target node can be determined based on at least one control point. The NURBS curve changes with the change of the node K, and the set of nodes K Ki: [K0, K1..., Kn..., Kn+m+1] is called a node vector. That is, the set of each target node constructs a node vector, and the node vector determines the influence range of each control point in the parameter space. The denser the distribution of the target nodes, the more localized the influence of the control points corresponding to the target nodes on the curve. Therefore, it can be seen that in the present disclosure, the node vector can be used to control the segmentation in the trajectory and the shape of each segment.
[0053] For the control points, each control point has a corresponding weight, which is used to characterize the "gravity" of the control point. The larger the weight, the closer the drawn curve is to the control point. In the present disclosure, the weight of the control point can be used to characterize the degree of attraction of the control point to the smoothed driving trajectory. The control point with a larger weight has a stronger attraction to the smoothed driving trajectory, making the driving trajectory closer to the vertex; the control point with a smaller weight has a relatively weaker influence on the smoothed driving trajectory, and the driving trajectory may be farther away from it.
[0054] The driving trajectory generated by using the node vector and the weight of the control point is the smoothed driving trajectory corresponding to the target driving trajectory.
[0055] In specific implementation, for any target driving trajectory, the N-order NURBS curve method can be used to smooth the target driving trajectory to obtain a smoothed driving trajectory. Wherein, N is the smoothing order of the above curve. The specific process of smoothing can be: first, multiple control points are selected from the target driving trajectory. For example, according to the preset point interval, multiple control points are sampled at equal intervals from the target driving trajectory. Then, according to the number of control points and the curve smoothing order, the target number of target nodes to be generated is determined, and then, according to the first coordinate information of each control point, the trajectory curvature at the control point in the target driving trajectory and other information, the weight of each control point and the target number of target nodes are determined, and then the node vector is composed of the target number of target nodes. Wherein, the node vector and the control point weight can be uniformly selected, that is, the weight of each control point can be set to the same value, for example, the weight of each control point is set to 1. Except for the different positions to be represented, the remaining information of each target node in the target vector can be the same.
[0056] S104-2: Generate a smoothed driving trajectory corresponding to the target driving trajectory according to the node vector and the weight of the control point.
[0057] In specific implementation, for any target driving trajectory, the trajectory can be drawn according to the node vector corresponding to the target driving trajectory and the weights of each control point, and the drawn trajectory is used as the smoothed driving trajectory corresponding to the target driving trajectory.
[0058] like Figure 2 , which is a schematic diagram of a comparison of historical driving trajectories before and after smoothing provided by an embodiment of the present disclosure, wherein: Figure 2 2-1 in the figure are the historical driving trajectories before smoothing. Figure 2 2-2 in the figure shows various smoothed driving trajectories obtained after trajectory smoothing using the trajectory smoothing method provided in the embodiment of the present disclosure.
[0059] In this way, in the process of trajectory smoothing, by selecting control points from the target driving trajectory, and based on the curve smoothing order and the first coordinate information of the control points, the node vector and control point weight used to draw the smoothed trajectory can be determined, and then the node vector and control point weight can be used to achieve reasonable smoothing of the target driving trajectory, and obtain an accurate and reasonable smoothed driving trajectory corresponding to the target driving trajectory. Furthermore, using the smoothed driving trajectory of other vehicles for decision planning of the self-vehicle can improve the accuracy and rationality of the decision.
[0060] In one embodiment, the above S101 may be implemented according to the following steps: S101 - 1 : Acquire initial trajectory information of each other vehicle detected by the own vehicle, where the initial trajectory information includes third coordinate information of multiple trajectory points in the own vehicle coordinate system.
[0061] Here, since the driving trajectory of the other vehicle is detected by the self-vehicle, the coordinates of each track point in the driving trajectory of the other vehicle detected by the self-vehicle are all coordinate information in the self-vehicle coordinate system, and these coordinate information are defined as the third coordinate information. That is, the third coordinate information includes the coordinate values of each track point in the historical driving trajectory of the other vehicle in the self-vehicle coordinate system.
[0062] The initial trajectory information is the information collected by the self-vehicle and related to the historical driving trajectory of the other vehicle, which may include the third coordinate information of each trajectory point in the historical driving trajectory of the other vehicle in the self-vehicle coordinate system, and the identifier corresponding to the historical driving trajectory of the other vehicle.
[0063] Exemplarily, during the driving process of the vehicle, the vehicle can detect at least one other vehicle (e.g., other vehicles located in front of the vehicle) that has an impact on the driving of the vehicle and determine the identification of each other vehicle. For each other vehicle, the vehicle continuously collects the third coordinate information of each track point in the historical driving track of the other vehicle under the coordinates of the vehicle and the identification of the other vehicle. Then, when it is determined that the track smoothing process is required, the third coordinate information of each track point of the other vehicle can be read in turn, and then the third coordinate information of each track point of the other vehicle and the identification of the other vehicle read are used as the initial track information of the other vehicle.
[0064] S101-2: According to the conversion relationship between the vehicle coordinate system and the northeast sky coordinate system, the third coordinate information of the plurality of trajectory points is converted into the first coordinate information.
[0065] S101 - 3: Use the first coordinate information of multiple trajectory points as driving trajectory information.
[0066] Here, the northeast sky coordinate system is the east-north-sky coordinate system, and the conversion relationship is used to represent the conversion relationship between the coordinates in the vehicle coordinate system and the coordinates in the east-north-sky coordinate system.
[0067] In specific implementation, the vehicle trajectory information can be obtained, and the conversion relationship between the vehicle coordinate system and the station center coordinate system can be determined according to the coordinate information of each trajectory point on the vehicle's driving trajectory included in the vehicle trajectory information in the east-north-sky coordinate system. Then, for each other's vehicle's initial trajectory information, the coordinate system conversion relationship between the vehicle coordinate system and the east-north-sky coordinate system can be used to convert the third coordinate information of each trajectory point in the other vehicle into the first coordinate information in the east-north-sky coordinate system in sequence according to the order of each trajectory point in the other vehicle. Then, the first coordinate information of each trajectory point in the other vehicle and the identification of the other vehicle can be used as the driving trajectory information of the other vehicle, and the trajectory formed by connecting the first coordinate information of each trajectory point can be used as the historical driving trajectory of the other vehicle.
[0068] In one embodiment, the step of screening control points in S104-1 may also be implemented according to the following steps: S104-1-1: According to the first coordinate information of each target trajectory point in the target driving trajectory, the previous trajectory point among the adjacent trajectory points whose spacing is less than a first distance is removed from the target driving trajectory in sequence; the first distance is less than a set distance.
[0069] Here, both the first distance and the set distance can be set based on experience, but the first distance should be smaller than the set distance. The first distance is set to filter adjacent trajectory points with a small spacing, thereby eliminating "burr" points in the trajectory and avoiding the overfitting effect caused by too close trajectory points on trajectory smoothing.
[0070] In specific implementation, for any target driving trajectory, the distance between each two adjacent trajectory points in the target driving trajectory can be determined in sequence according to the first coordinate information of each target trajectory point in the target driving trajectory, and it is determined whether the distance is less than the first distance. If so, the previous trajectory point of the two trajectory points is removed from the target driving trajectory; if not, the trajectory point is retained.
[0071] S104-1-2: Filter out a plurality of control points from the reserved track points in the target driving track according to the track length of the target driving track.
[0072] Here, the retained trajectory points are the trajectory points in the target driving trajectory that are not eliminated after the trajectory points in the target driving trajectory are eliminated using the first distance.
[0073] In a specific implementation, after the track points in the target driving track are eliminated by using the first distance, the distance between each two adjacent retained track points can be determined according to the first coordinate information of each retained track point in the target driving track, and the track length of the target driving track can be determined according to the sum of the distances between each two adjacent retained track points. Then, the control point selection method is determined by using the track length, and multiple control points are screened out from the retained track points by using the control point selection method.
[0074] Exemplarily, when the trajectory length is greater than a preset length, the control point selection method is determined to sample every m retained trajectory points, where m is a preset value greater than 1; when the trajectory length is not greater than a preset length, the control point selection method is determined to use each retained trajectory point as a control point.
[0075] Alternatively, the ratio of the trajectory length to the preset control point interval may be rounded up, and the obtained value may be used as the number of sampling interval points. The retained trajectory points may be selected according to the number of sampling interval points to obtain various control points.
[0076] In one embodiment, the above S104-1-2 can be implemented according to the following steps: When the trajectory length is greater than or equal to the first length, selecting a plurality of control points from the reserved trajectory points in the target driving trajectory according to the first interval value; When the trajectory length is less than or equal to the second length, all the reserved trajectory points in the target driving trajectory are used as control points; the first length is greater than the second length; When the track length is between the second length and the first length, a plurality of control points are screened out from the retained track points in the target driving track according to a second interval value; the second interval value is smaller than the first interval value.
[0077] Here, the first length, the second length, the first interval value, and the second interval value can all be set based on experience, and are not specifically limited here, but the first length is greater than the second length, and the second interval value is less than the first interval value. For example, the first length can be 50m, the second length can be 20m, the first interval value can be 3, and the second interval value can be 2.
[0078] When the trajectory length is greater than or equal to the first length, it means that the target driving trajectory is longer and the number of retained trajectory points is larger. In this case, a larger uniform sampling interval can be used for equal interval sampling. Therefore, the first interval value can be used as the uniform sampling interval, and each retained trajectory point in the target driving trajectory can be uniformly sampled according to the first interval value and the order of each retained trajectory point in the target driving trajectory to obtain multiple control points.
[0079] When the trajectory length is less than or equal to the second length, it means that the target driving trajectory is very short and the number of retained trajectory points is very small. In order to ensure that there are enough control points for smoothing, each retained trajectory point in the target driving trajectory can be used as a control point. That is, each retained trajectory point in the target driving trajectory is uniformly sampled using an interval value of 0 to obtain each control point.
[0080] When the trajectory length is between the second length and the first length, it means that the target driving has a medium length and a medium number of retained trajectory points. At this time, if a larger sampling interval is used for sampling, the screened control points may be insufficient for smoothing. If all the retained trajectory points are directly used as control points, there may be a problem of smooth transition due to too many control points. Therefore, a second interval value smaller than the first interval value can be used as a uniform sampling interval, and each retained trajectory point in the target driving trajectory can be uniformly sampled according to the second interval value and the order of each retained trajectory point in the target driving trajectory to obtain multiple control points.
[0081] Further, after selecting multiple control points from the target driving trajectory, it can also be determined whether the number of control points corresponding to the target driving trajectory is less than or equal to the curve smoothing order. If so, it means that there is a problem in the control point screening process, and the number of control points is not enough for smoothing. It is necessary to make a bottom-up here, so the disclosed embodiment sets a bottom-up strategy, that is, when the number of multiple control points is less than or equal to the curve smoothing order, each reserved track point in the target driving trajectory is used as a control point, and the control point is updated to ensure that the target driving trajectory corresponds to a sufficient number of control points for trajectory smoothing. Of course, if the number of control points is still less than or equal to the curve smoothing order after each reserved track point in the target driving trajectory is used as a control point, the target driving trajectory can be directly eliminated. It is understandable that for each target driving trajectory eliminated during the smoothing process, when the smoothed driving trajectory is subsequently visualized, the eliminated target driving trajectories will no longer be visualized, or the target driving trajectory can be directly visualized.
[0082] In one embodiment, with respect to the step of “using the first driving trajectory and the third driving trajectory as the target driving trajectory” in S103, trajectories whose trajectory length is less than the second distance in the first driving trajectory and the third driving trajectory may be eliminated, and the remaining trajectory may be used as the target driving trajectory, wherein the second distance is greater than the first distance.
[0083] Here, the second distance can be set based on experience and is not specifically limited here, but the second distance must be greater than the first distance and less than the above-mentioned second length.
[0084] In specific implementation, after obtaining each first driving trajectory and the third driving trajectory, in order to further ensure that the trajectory to be subsequently smoothed has a sufficient length, for any first driving trajectory and any third driving trajectory, it can also be determined whether the trajectory length of each driving trajectory is greater than or equal to the second distance. If so, it means that the driving trajectory has a certain length and can be smoothed, then the driving trajectory is retained as the target driving trajectory and the above-mentioned smoothing processing step corresponding to 104 is performed for the target driving trajectory. If not, that is, when the trajectory length of the driving trajectory is less than the second distance, it means that the driving trajectory is a very short trajectory, which may be a trajectory formed by a stationary or slowly moving other vehicle at a traffic light intersection, roadside, etc. Such other vehicles often have no effect on the driving and intelligent decision-making of the vehicle, so the driving trajectory can be directly eliminated to reduce the calculation amount of trajectory smoothing and save computing resources.
[0085] In one embodiment, S104-2 may be implemented according to the following steps: S104-2-1: According to the node vector and the node weight, at least two new target nodes are inserted at equal intervals between every two target nodes, and the node vector is updated using each inserted target node.
[0086] In a specific implementation, in order to ensure the smoothness and uniformity of the output smoothed driving trajectory, after obtaining the node vector and the weight of the control point, node difference processing can also be performed to increase the number of nodes.
[0087] Specifically, for each target node obtained, at least two new target nodes may be uniformly inserted (i.e., inserted at equal intervals) between every two target nodes, and then, each target node before the insertion and each newly inserted target node are used to combine a new node vector, and the new node vector is used to update the node vector composed of each target node before the insertion. For example, the node vector composed of each target node before the insertion is replaced by the new node vector.
[0088] For example, two new target nodes are inserted between target node 1 and target node 2. The two newly inserted target nodes are respectively recorded as node 3 and node 4. The parameters of node 3 can be the parameters corresponding to one-third of the distance from the parameters of target node 1 to the parameters of target node 2, and the parameters of node 4 can be the parameters corresponding to two-thirds of the distance from the parameters of target node 1 to the parameters of target node 2.
[0089] In this way, the trajectory of the other vehicle is smoothed by using a NURBS curve of a curve smoothing order, and optimization processing is also performed in the process of control point selection and interpolation output, such as initial distance screening of control points, uniform sampling method for selecting control points, and interpolation output of new target nodes. This can enable the output of a smooth trajectory corresponding to the other vehicle, avoid the "burr" phenomenon in the trajectory of the other vehicle, and reduce the noise in the trajectory of the other vehicle.
[0090] S104-2-2: Draw a non-uniform rational B-spline curve based on the weight of each control point and the updated node vector.
[0091] Here, the non-uniform rational B-spline curve is a NURBS curve.
[0092] Exemplarily, a NURBS curve method may be used to draw a NURBS curve according to the weight of each control point and the updated node vector.
[0093] S104-2-3: Convert the first coordinate information of each curve point in the non-uniform rational B-spline curve into the second coordinate information in the longitude and latitude coordinate system.
[0094] Here, the curve point may be a key position point in the non-uniform rational B-spline curve, the curve point matches the target node, and the position of the curve point is determined by the position of the target node.
[0095] In specific implementation, after drawing the NURBS curve, each curve point in the NURBS curve and the first coordinate information of the curve point in the ENU coordinate system can be determined, and then the first coordinate information of each curve point can be converted into the second coordinate information in the LLA coordinate system according to the conversion relationship between the ENU coordinate system and the longitude and latitude coordinate system (i.e., the LLA coordinate system).
[0096] S104-2-4: Generate a smoothed driving trajectory corresponding to the target driving trajectory based on the second coordinate information.
[0097] Exemplarily, a curve in the LLA coordinate system is drawn according to the second coordinate information of each curve point, and the curve is used as the smoothed driving trajectory corresponding to the target driving trajectory. Afterwards, the generated smoothed driving trajectory can be visualized and output on the display end.
[0098] In one embodiment, after executing S104-2, the following steps may also be executed: Step 1: For any target curve point except the first curve point and the last curve point in the smoothed driving trajectory, determine the curvature of the target curve point according to the second coordinate information of the curve points adjacent to the target curve point in front and behind in the longitude and latitude high coordinate system.
[0099] Here, the curve points in the smoothed driving trajectory are points at each second coordinate information in the longitude and latitude coordinate system when drawing the smoothed driving trajectory.
[0100] In the specific implementation, the curvature calculation method of three adjacent discrete points is adopted to calculate the curvature of the curve points in the smoothed driving trajectory. Among them, since there is no adjacent curve point before the first curve point and there is no adjacent curve point after the last curve point, the curvature of the first curve point and the last curve point cannot be calculated. The target curve point is any curve point other than the first curve point and the last curve point among the curve points included in the smoothed driving trajectory. For each target curve point in the smoothed driving trajectory, the curvature of the target curve point can be calculated according to the second coordinate information of the curve points adjacent to the target curve point before and after in the longitude and latitude high coordinate system. In this way, the curvature of each target curve point except the first curve point and the last curve point in the smoothed driving trajectory can be obtained.
[0101] Step 2: Determine the curvature difference between every two adjacent target curve points according to the curvature of the target curve point.
[0102] Exemplarily, the curvature difference between every two adjacent target curve points may be calculated according to the curvature of every two adjacent target curve points.
[0103] Step 3: Sort the curvature differences in order from small to large, and use the curvature difference at the target percentile as the smoothness evaluation index of the smoothed driving trajectory.
[0104] Here, the target percentile can be set based on experience, for example, the target percentile can be the 90th percentile, the 95th percentile, etc. The 95th percentile means the 95th percentile. The smoothness evaluation index is used to indicate the overall smoothness of the smoothed driving trajectory.
[0105] For example, after obtaining the curvature difference between every two adjacent target curve points, the curvature differences can be sorted in ascending order to obtain the sorting order of each curvature difference. Then, the curvature difference with the 95th percentile in the sorting order can be used as the smoothness evaluation index of the smoothed driving trajectory.
[0106] In this way, the sorting of curvature differences and taking the value at the target percentile as the smoothness evaluation index can ideally represent the trajectory smoothness of the smoothed driving trajectory.
[0107] like Figure 3 FIG. 1 is a schematic diagram of a specific implementation process of a trajectory smoothing method provided by an embodiment of the present disclosure, which may include the following steps: S301: Obtain the trajectory information of the own vehicle and the initial trajectory information of each other vehicle.
[0108] S302: Data preprocessing to obtain each target driving trajectory.
[0109] Among them, the data preprocessing step may include converting the third coordinate information of each trajectory point in the initial trajectory information of the other vehicle into the first coordinate information in the east-north-sky coordinate system according to the trajectory information of the own vehicle, and dividing the historical driving trajectory in which the distance between adjacent trajectory points is not less than the set distance.
[0110] S303: Short-term trajectory filtering.
[0111] That is, the target driving trajectories whose trajectory length is less than the second distance are eliminated from the target driving trajectories.
[0112] S304: NURBS curve smoothing.
[0113] The specific implementation process of this step can refer to the specific implementation process of S104-1 and S104-2 above, which will not be repeated here.
[0114] S305: Output a smoothness evaluation index.
[0115] The specific implementation process of this step can refer to the specific implementation process of the above steps 1 to 3, which will not be repeated here.
[0116] Those skilled in the art will appreciate that, in the above method of specific implementation, the order in which the steps are written does not imply a strict execution order and does not constitute any limitation on the implementation process. The specific execution order of the steps should be determined by their functions and possible internal logic.
[0117] In addition, an embodiment of the present disclosure further provides a vehicle, wherein the vehicle includes a controller, wherein the controller includes: a memory for storing machine-readable instructions; and The processor is used to call and execute the machine-readable instructions stored in the memory to implement the steps of the trajectory smoothing method described in the above embodiments.
[0118] Based on the same technical concept, the embodiment of the present application also provides a computer device. Figure 4 FIG. 1 is a schematic diagram of a computer device provided in an embodiment of the present application, including: Processor 401, memory 402 and bus 403. The memory 402 stores machine-readable instructions executable by the processor 401, and the processor 401 is used to execute the machine-readable instructions stored in the memory 402. When the machine-readable instructions are executed by the processor 401, the processor 401 performs the following steps: S101: Acquire driving trajectory information of each other vehicle detected by the self-vehicle; the driving trajectory information includes first coordinate information of multiple trajectory points in the historical driving trajectory of the other vehicle; S102: Filter out a first driving trajectory and a second driving trajectory from each historical driving trajectory according to the first coordinate information of the multiple trajectory points; the distance between each adjacent trajectory point in the first driving trajectory is less than the set distance, and the distance between at least two adjacent trajectory points in the second driving trajectory is greater than or equal to the set distance; S103: Use at least two adjacent trajectory points in the second driving trajectory whose distance is greater than or equal to the set distance to divide the second driving trajectory into at least two third driving trajectories, and use the first driving trajectory and the third driving trajectory as target driving trajectories; and S104: Smooth the target driving trajectory to obtain a smoothed driving trajectory.
[0119] The above-mentioned memory 402 includes internal memory 4021 and external memory 4022; the memory 4021 here is also called internal memory, which is used to temporarily store the calculation data in the processor 401, as well as the data exchanged with the external memory 4022 such as the hard disk. The processor 401 exchanges data with the external memory 4022 through the internal memory 4021. When the computer device is running, the processor 401 communicates with the memory 402 through the bus 403, so that the processor 401 executes the execution instructions mentioned in the above-mentioned method embodiment.
[0120] The present disclosure also provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the trajectory smoothing method described in the above method embodiment are executed. The storage medium can be a volatile or non-volatile computer-readable storage medium.
[0121] The computer program product of the trajectory smoothing method provided in the embodiments of the present disclosure includes a computer-readable storage medium storing program code, and the instructions included in the program code can be used to execute the steps of the trajectory smoothing method described in the above method embodiments. For details, please refer to the above method embodiments, which will not be repeated here.
[0122] The computer program product may be implemented in hardware, software or a combination thereof. In one optional embodiment, the computer program product is embodied as a computer storage medium, and in another optional embodiment, the computer program product is embodied as a software product, such as a software development kit (SDK).
[0123] Those skilled in the art can clearly understand that, for the convenience and simplicity of description, the specific working process of the system and device described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here. In the several embodiments provided in the present disclosure, it should be understood that the disclosed system, device and method can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interfaces, and the indirect coupling or communication connection of the device or unit can be electrical, mechanical or other forms.
[0124] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
[0125] In addition, each functional unit in each embodiment of the present disclosure may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.
[0126] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a non-volatile computer-readable storage medium that is executable by a processor. Based on this understanding, the technical solution of the present disclosure, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present disclosure. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.
[0127] If the technical solution of this application involves personal information, the product using the technical solution of this application has clearly informed the personal information processing rules and obtained the individual's voluntary consent before processing the personal information. If the technical solution of this application involves sensitive personal information, the product using the technical solution of this application has obtained the individual's separate consent before processing the sensitive personal information, and at the same time meets the "explicit consent" requirement. For example, on personal information collection devices such as cameras, clear and prominent signs are set to inform that the personal information collection scope has been entered and personal information will be collected. If the individual voluntarily enters the collection scope, it is deemed that he or she agrees to the collection of his or her personal information; or on the device that processes personal information, the personal information processing rules are notified by obvious signs / information, and the individual's authorization is obtained through pop-up information or by asking the individual to upload his or her personal information; among them, the personal information processing rules may include information such as the personal information processor, the purpose of personal information processing, the processing method, and the type of personal information processed.
[0128] Finally, it should be noted that the above-described embodiments are only specific implementation methods of the present disclosure, which are used to illustrate the technical solutions of the present disclosure, rather than to limit them. The protection scope of the present disclosure is not limited thereto. Although the present disclosure is described in detail with reference to the aforementioned embodiments, ordinary technicians in the field should understand that any technician familiar with the technical field can still modify the technical solutions recorded in the aforementioned embodiments within the technical scope disclosed in the present disclosure, or can easily think of changes, or make equivalent replacements for some of the technical features therein; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present disclosure, and should be included in the protection scope of the present disclosure. Therefore, the protection scope of the present disclosure shall be based on the protection scope of the claims.
Claims
1. A trajectory smoothing method, characterized in that: include: Obtain the driving trajectory information of each other vehicle detected by the self-vehicle; The driving track information includes first coordinate information of a plurality of track points in the historical driving track of the other vehicle; Filtering out a first driving trajectory and a second driving trajectory from each of the historical driving trajectories according to the first coordinate information of the plurality of trajectory points; The distance between each adjacent track point in the first driving track is less than a set distance, and the distance between at least two adjacent track points in the second driving track is greater than or equal to the set distance; Using at least two adjacent track points in the second driving track whose distance is greater than or equal to a set distance, the second driving track is divided into at least two third driving tracks, and the first driving track and the third driving track are used as target driving tracks; The target driving trajectory is smoothed to obtain a smoothed driving trajectory.
2. The method according to claim 1, characterized in that The step of smoothing the target driving trajectory to obtain a smoothed driving trajectory includes: A plurality of control points are selected from the target driving trajectory, and a weight of each control point and a node vector composed of target nodes corresponding to the plurality of control points are determined according to a curve smoothing order and first coordinate information of the control point; each target node is determined according to at least one control point, and the node vector is used to control segments in the trajectory and the shape of each segment; A smoothed driving trajectory corresponding to the target driving trajectory is generated according to the node vector and the weight of the control point.
3. The method according to claim 2, characterized in that The step of selecting a plurality of control points from the target driving trajectory includes: According to the first coordinate information of each target trajectory point in the target driving trajectory, the previous trajectory point among the adjacent trajectory points whose interval is less than a first distance is removed from the target driving trajectory in sequence; the first distance is less than the set distance; According to the track length of the target driving track, a plurality of control points are screened out from the reserved track points in the target driving track.
4. The method according to claim 3, characterized in that According to the track length of the target driving track, a plurality of control points are screened out from the reserved track points in the target driving track, including: When the trajectory length is greater than or equal to the first length, selecting a plurality of control points from the reserved trajectory points in the target driving trajectory according to the first interval value; In the case where the trajectory length is less than or equal to the second length, all the reserved trajectory points in the target driving trajectory are used as the control points; the first length is greater than the second length; When the trajectory length is between the second length and the first length, a plurality of control points are screened out from the retained trajectory points in the target driving trajectory according to a second interval value; the second interval value is smaller than the first interval value.
5. The method according to claim 3, characterized in that: After selecting a plurality of control points from the target driving trajectory, the method further includes: When the number of the plurality of control points is less than or equal to the curve smoothing order, all the reserved track points in the target driving track are used as the control points.
6. The method according to claim 2, characterized in that The step of generating a smoothed driving trajectory corresponding to the target driving trajectory according to the node vector and the weight of the control point includes: According to the node vector, at least two new target nodes are inserted at equal intervals between every two target nodes, and the node vector is updated using each inserted target node; Drawing a non-uniform rational B-spline curve according to the weight of each control point and the updated node vector; Converting the first coordinate information of each curve point in the non-uniform rational B-spline curve into the second coordinate information in the longitude and latitude coordinate system; A smoothed driving trajectory corresponding to the target driving trajectory is generated according to the second coordinate information.
7. The method according to claim 2, characterized in that After generating the smoothed driving trajectory corresponding to the target driving trajectory according to the node vector and the weight of the control point, the method further includes: For any target curve point except the first curve point and the last curve point in the smoothed driving trajectory, determine the curvature of the target curve point according to the second coordinate information of the curve points adjacent to the target curve point in front and behind in the longitude and latitude coordinate system; Determining the curvature difference between every two adjacent target curve points according to the curvature of the target curve point; The curvature differences are sorted in order from small to large, and the curvature difference at the target percentile in the sorting order is used as the smoothness evaluation index of the smoothed driving trajectory.
8. The method according to any one of claims 1 to 7, characterized in that The taking the first driving trajectory and the third driving trajectory as target driving trajectories includes: Trajectories whose trajectory lengths are less than the second distance in the first driving trajectory and the third driving trajectory are eliminated, and the remaining trajectory is used as the target driving trajectory.
9. The method according to any one of claims 1 to 7, characterized in that The step of obtaining the driving trajectory information of each other vehicle detected by the vehicle includes: Acquire initial trajectory information of each other vehicle detected by the self-vehicle, wherein the initial trajectory information includes third coordinate information of multiple trajectory points in the self-vehicle coordinate system; According to the conversion relationship between the vehicle coordinate system and the northeast sky coordinate system, the third coordinate information of the plurality of trajectory points is converted into the first coordinate information; The first coordinate information of multiple trajectory points is used as the driving trajectory information.
10. A vehicle, characterized in that: A controller is included, the controller comprising: a memory for storing machine-readable instructions; and A processor, configured to call and execute the machine-readable instructions stored in the memory to implement the steps of the trajectory smoothing method according to any one of claims 1 to 9.
11. A computer device, characterized in that: include: A processor and a memory, wherein the memory stores machine-readable instructions executable by the processor, and the processor is used to execute the machine-readable instructions stored in the memory. When the machine-readable instructions are executed by the processor, the processor executes the steps of the trajectory smoothing method according to any one of claims 1 to 9.
12. A computer program product, which, when executed by a computer device, performs the steps of the trajectory smoothing method according to any one of claims 1 to 9.
Citation Information
Patent Citations
Method and system for replaying vehicle historical traveling track
CN103927795A
Driving data generation method and device based on fitting track, equipment and medium
CN113345228A
Vehicle travel path identification method in urban dense road network environment
CN115578860A
Method and system for automatic driving of off-road vehicle following set target and medium
CN117032241A
Method for consolidating a trajectory from trajectory fragments.
FR3095858A1
