Intersection turning trajectory planning method and device, vehicle and storage medium

Abstract

The application discloses a kind of intersection turning trajectory planning method, device, vehicle and storage medium, it is related to intelligent driving technical field, method includes: obtaining navigation map information, vehicle positioning information and vehicle perception information, vehicle perception information is used to characterize the environmental information of the target intersection where the vehicle is located, target intersection is the intersection of the road where the vehicle is currently located;Based on navigation map information, generate the first turning trajectory corresponding to the target intersection, navigation map information at least includes the information of multiple key points in the preset range of the target intersection;Based on navigation map information, vehicle positioning information and vehicle perception information, adjust the first turning trajectory, obtain the second turning trajectory.The application makes the vehicle in various turning intersections can generate the first turning trajectory according to navigation map information, then adjust and obtain suitable second turning trajectory in combination with vehicle perception information, realize the automatic turning traffic of vehicle in intersection, improve the robustness of turning traffic.

Description

Technical Field

[0001] This application relates to the field of intelligent driving technology, and more specifically, to a method, apparatus, vehicle, and computer-readable storage medium for intersection turning trajectory planning. Background Technology

[0002] With the development and application of intelligent driving systems in urban scenarios, autonomous driving technology also needs to adapt to various complex urban driving conditions, such as turning at urban intersections. The working principle of turning at urban intersections is as follows: the vehicle selects a turning lane before the intersection, determines whether to turn based on the lane type and traffic light status, and during the turning process, controls the vehicle to complete the turn and enter the corresponding lane based on the turning curvature information provided by the navigation system, as well as information such as lane lines and road edge boundaries.

[0003] Currently, most turning schemes used in autonomous driving systems still rely on high-precision maps to provide the information needed for turning behavior. However, due to the time-sensitivity of high-precision map updates, erroneous data can easily be obtained. Under the influence of erroneous data, it is difficult to complete turning, which seriously affects the driving experience and safety. Summary of the Invention

[0004] This application proposes a method, device, vehicle, and storage medium for intersection turning trajectory planning to improve the above-mentioned deficiencies.

[0005] In a first aspect, embodiments of this application provide a method for planning a turning trajectory at an intersection. The method is applied to a vehicle and includes: acquiring navigation map information, vehicle positioning information, and vehicle perception information, wherein the vehicle perception information is used to characterize the environmental information of the target intersection where the vehicle is located, and the target intersection is the intersection where the vehicle is currently located; generating a first turning trajectory corresponding to the target intersection based on the navigation map information, wherein the navigation map information includes at least information on multiple key points within a preset range of the target intersection; and adjusting the first turning trajectory based on the navigation map information, vehicle positioning information, and vehicle perception information to obtain a second turning trajectory.

[0006] Secondly, this application also provides an intersection turning trajectory planning device, which is applied to a vehicle and includes: an information acquisition module for acquiring navigation map information, vehicle positioning information, and vehicle perception information, wherein the vehicle perception information is used to characterize the environmental information of the target intersection where the vehicle is located, and the target intersection is the intersection where the vehicle is currently located; a trajectory generation module for generating a first turning trajectory corresponding to the target intersection based on the navigation map information, wherein the navigation map information includes at least information on multiple key points within a preset range of the target intersection; and a trajectory adjustment module for adjusting the first turning trajectory based on the navigation map information, vehicle positioning information, and vehicle perception information to obtain a second turning trajectory.

[0007] Thirdly, embodiments of this application also provide a vehicle, including: a perception sensor for acquiring vehicle perception information; one or more processors; a memory; one or more applications, wherein the one or more applications are stored in the memory and configured to be executed by the one or more processors, and the one or more applications are configured to perform the methods described above.

[0008] Fourthly, embodiments of this application also provide a computer-readable storage medium storing processor-executable program code, which, when executed by the processor, causes the processor to perform the above-described method.

[0009] Therefore, this application provides a method, apparatus, vehicle, and storage medium for planning turning trajectories at intersections. The method is applied to a vehicle and includes: acquiring navigation map information, vehicle positioning information, and vehicle perception information, wherein the vehicle perception information is used to characterize the environmental information of the target intersection where the vehicle is located, and the target intersection is the intersection of the road where the vehicle is currently located; generating a first turning trajectory corresponding to the target intersection based on the navigation map information, wherein the navigation map information includes at least information on multiple key points within a preset range of the target intersection; and adjusting the first turning trajectory based on the navigation map information, vehicle positioning information, and vehicle perception information to obtain a second turning trajectory. Therefore, this application generates a first turning trajectory for passing through a target intersection using navigation map information, and adjusts the first turning trajectory based on the first turning trajectory using navigation map information, vehicle positioning information, and vehicle perception information to obtain a more accurate second turning trajectory that better reflects the actual traffic conditions of the road. Thus, the method provided in this application can generate a first turning trajectory based on navigation map information at different turning intersections, and then adjust and obtain a suitable second turning trajectory by combining vehicle perception information, enabling automatic turning passage of vehicles at intersections and improving the robustness of turning passage in autonomous driving.

[0010] Other features and advantages of the embodiments of this application will be set forth in the following description, and will be apparent in part from the description, or may be learned by practicing the embodiments of this application. The objects and other advantages of the embodiments of this application may be realized and obtained by means of the structures particularly pointed out in the written description, claims, and drawings. Attached Figure Description

[0011] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0012] Figure 1 A schematic diagram of a vehicle hardware environment applicable to embodiments of this application is shown.

[0013] Figure 2 A flowchart of an intersection turning trajectory planning method proposed in an embodiment of this application is shown.

[0014] Figure 3 This diagram illustrates a turning trajectory of a vehicle at a target intersection, as provided in an embodiment of this application.

[0015] Figure 4 A flowchart of another intersection turning trajectory planning method provided in an embodiment of this application is shown.

[0016] Figure 5 A flowchart of another intersection turning trajectory planning method provided in an embodiment of this application is shown.

[0017] Figure 6 A structural block diagram of an intersection turning trajectory planning device according to an embodiment of this application is shown.

[0018] Figure 7 A structural block diagram of a computer-readable storage medium provided in an embodiment of this application is shown. Detailed Implementation

[0019] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, and not all of them. The components of the embodiments of the present application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely represents selected embodiments of the present application. All other embodiments obtained by those skilled in the art based on the embodiments of the present application without inventive effort are within the scope of protection of the present application.

[0020] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this application, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0021] Please see Figure 1 , Figure 1 A schematic diagram of a vehicle hardware environment applicable to an embodiment of this application is shown. The vehicle 100 includes a processor 111, a memory 112, and a sensing sensor 113.

[0022] The processor 111 may be a microcontroller unit (MCU), which has a built-in memory 112 storing programs that can execute the contents described in the following embodiments. The processor 111 can execute the programs stored in the memory 112. Furthermore, the processor 111 is connected to the cloud, and data interaction can occur between the processor 111 and the cloud. This allows the processor 111 to upload relevant data acquired by the sensing sensor 113 to the cloud for further processing, and the processor 111 can also receive information from the cloud to perform related operations.

[0023] The processor 111 may include one or more processing cores. The processor 111 connects to various parts of the vehicle 100 via various interfaces and lines, and performs various functions and processes data of the vehicle 100 by running or executing instructions, programs, code sets, or instruction sets stored in the memory 112, and by calling data stored in the memory 112. Optionally, the processor 111 may be implemented using at least one of the following hardware forms: a Neural Network Processing Unit (NPU), a Digital Signal Processing Unit (DSP), a Field-Programmable Gate Array (FPGA), or a Programmable Logic Array (PLA). The processor 111 may integrate one or a combination of several of the following: a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), a Neural Network Processing Unit (NPU), and a modem. The CPU primarily handles the operating system, user interface, and applications; the GPU is responsible for rendering and drawing the displayed content; the NPU is responsible for processing multimedia data such as video and images; and the modem is used for wireless communication. It is understandable that the aforementioned modem may not be integrated into the processor 111, but may be implemented using a separate communication chip.

[0024] The memory 112 may include random access memory (RAM), read-only memory (ROM), and double data rate synchronous dynamic random access memory (DDR). The memory 112 can be used to store instructions, programs, code, code sets, or instruction sets. The memory 112 may include a program storage area and a data storage area. The program storage area may store instructions for implementing an operating system, instructions for implementing at least one function (such as a turning trajectory planning function), instructions for implementing the various method embodiments described below, etc. The data storage area may also store data generated by the vehicle 100 during use (such as perception information), etc.

[0025] The perception sensor 113 may include sensors such as an onboard camera and millimeter-wave radar, used to detect and output perception information about lanes and intersections around the vehicle. For example, the onboard camera may include a front-view camera and a side-view camera, used to acquire real-time images of the vehicle's surroundings and identify perception information such as lane line information, curb information, traffic light information, and obstacle information from the images; the millimeter-wave radar may include a front millimeter-wave radar and an angular millimeter-wave radar, used to measure the distance between the vehicle and surrounding objects, thereby combining the information acquired by the camera to determine the specific location of the perception information.

[0026] Please see Figure 2 , Figure 2 This invention illustrates a flowchart of a method for planning turning trajectories at intersections, as provided in an embodiment of this application. The method can be applied to, for example... Figure 1 On the vehicle shown, specifically, the method includes the following steps:

[0027] S110: Obtain navigation map information, vehicle positioning information, and vehicle perception information. The vehicle perception information is used to characterize the environmental information of the target intersection where the vehicle is located. The target intersection is the intersection where the vehicle is currently waiting to turn on the road.

[0028] As one implementation method, navigation map information can be obtained from SD (Standard Definition) Map, a standard definition navigation map. Generally, SD navigation stores road-level elements with a resolution of around 5m-10m. Further, after the driver sets the navigation start point and destination, SD navigation can generate a navigation route based on these, thereby obtaining all intersections requiring turns along the route as turning points. Therefore, navigation map information can include: lane information, intersection turning signals, intersection turning action signals, and turning key point information. Lane information refers to the lane attributes of each lane related to the turning point output by the navigation map. Based on this lane information, it is possible to obtain the lanes available for turning before the turning point and the lane to continue driving into after completing the turning action at the turning point. Intersection turning signals refer to the signals issued by the navigation map in advance before the vehicle arrives at the turning point to indicate whether a left or right turn is permitted at the upcoming intersection. Generally, when an intersection turning signal is issued... The vehicle is far from the stop line at the intersection where it is about to turn. The stop line at the intersection can refer to the stop line corresponding to the lane before entering the intersection. The intersection turning signal refers to the turning instruction signal issued by the navigation map when the vehicle arrives at the intersection. Generally, the vehicle is close to the stop line at the intersection when the intersection turning signal is issued. Key points refer to the key points that provide route decision-making during the vehicle's turn at the intersection. If the intersection where the vehicle is currently located is taken as the target intersection, the turning key point information refers to the information of multiple key points within the preset range of the target intersection.

[0029] As one implementation method, vehicle positioning information can be obtained using an onboard Global Positioning System (GPS) to pinpoint the vehicle's exact location. It is understood that based on vehicle positioning information, the vehicle's specific location throughout the entire driving process and its lane can be determined, thus providing a positional reference for the turning trajectory planning process.

[0030] As one implementation method, vehicle perception information can be acquired based on perception sensors mounted on the vehicle. Specifically, these sensors can include cameras, lidar, or millimeter-wave radar. Cameras capture images of the vehicle's surrounding environment, while lidar or millimeter-wave radar detects the distance between objects around the vehicle and the vehicle itself. Thus, the perception sensors can acquire information about the intersection where the vehicle is about to turn in real time and accurately. Therefore, by taking the intersection where the vehicle is currently located as the target intersection, the vehicle perception information can be used to characterize the environmental information of the target intersection. Furthermore, the vehicle perception information can include: the target intersection and surrounding objects such as lane lines, stop lines, zebra crossings, curbs, obstacles, traffic lights, and ground markings, as well as the relative position information of these objects and the vehicle.

[0031] As one implementation method, navigation map information, vehicle positioning information, and vehicle perception information can be continuously acquired at preset time intervals to ensure that the latest navigation map information, vehicle positioning information, and vehicle perception information are available during vehicle operation. In particular, the vehicle perception information can reflect the latest road conditions around the vehicle, thereby ensuring that the planned turning trajectory can be adjusted based on the latest vehicle perception information in subsequent methods.

[0032] S120: Based on the navigation map information, generate a first turning trajectory corresponding to the target intersection, wherein the navigation map information includes at least information on multiple key points within a preset range of the target intersection.

[0033] As one implementation method, according to the description of navigation map information in step S110, the navigation map information includes at least turning key point information, that is, information on multiple key points within a preset range of the target intersection. Therefore, based on the turning key point information related to the target intersection in the navigation map information, a first turning trajectory corresponding to the target intersection can be generated. Furthermore, the navigation map information may also include lane information, intersection turning signals, intersection turning action signals, etc. Therefore, the lane information and intersection turning signals in the navigation map information can be combined to determine whether the target intersection has turning conditions. Thus, before the vehicle arrives at the target intersection, it can be known that the vehicle needs to turn at the target intersection. Then, based on the intersection turning action signals in the navigation map information, it can be known that at the time the navigation map issues the intersection turning action signal, the vehicle is close to the stop line of the target intersection. Finally, based on the turning key point information related to the target intersection in the navigation map information, a first turning trajectory corresponding to the target intersection can be generated.

[0034] As one implementation method, the method for generating the first turning trajectory corresponding to the target intersection based on navigation map information can be: before the vehicle is about to reach the stop line position of the target intersection, generate the first turning trajectory based on the turning key point information in the navigation map information.

[0035] Specifically, information on multiple key points within a preset range of the target intersection can be obtained based on navigation map information. These key points can refer to road intersections, the start and end points of roads, and a series of points describing the road shape. By connecting multiple key points, the curved sections of the road can be mapped to a series of adjacent straight line segments. The preset range refers to a circular area centered on the center point of the target intersection, covering the entire road surface of the target intersection. Then, based on a vehicle kinematics model, the sub-trajectory between each pair of adjacent key points is smoothed. The vehicle kinematics model primarily focuses on the changes in the vehicle's position, speed, acceleration, and other motion states over time, as well as the relationships between these states. Before smoothing the sub-trajectory of the straight line segments, the desired shape and curvature changes of the smoothed trajectory can be determined based on the vehicle's driving scenario and task. Then, common curve fitting methods such as polynomial fitting and spline curve fitting are used to smooth the sub-trajectory between each pair of key points into a suitable curved trajectory. Finally, each smoothed sub-trajectory is combined as the first turning trajectory of the target intersection.

[0036] Further, please refer to Figure 3 , Figure 3 This illustration shows a schematic diagram of a vehicle's turning trajectory at a target intersection, as depicted in an embodiment of this application. For ease of understanding, Figure 3 Let's take a vehicle turning left at the target intersection as an example for explanation, and Figure 3 The diagram shows key points A, B, C, and D at the target intersection. It can be seen that connecting every two key points yields a series of straight line segments, or sub-trajectory routes, that can be roughly used to plan the vehicle's turning trajectory. Then, by smoothing the sub-trajectory routes (dashed lines) between every two key points, the first turning trajectory (solid lines) is obtained. In some implementations, to reduce computational load, smoothing can be performed only on the sub-trajectory routes formed by key points within the target intersection, without smoothing the sub-trajectory routes formed by key points in lanes related to the target intersection. Figure 3 As shown, smoothing is performed only on the sub-trajectory route formed by key points B and C located at the target intersection.

[0037] As one implementation method, vehicles can make left or right turns at the target intersection. Since there are certain differences in the traffic rules that need to be followed when a vehicle makes a left turn and when it makes a right turn, the method shown in this embodiment can be further optimized in practical applications based on the differences between left and right turns. This will not be elaborated here.

[0038] Therefore, in this embodiment, by utilizing the information of multiple key points within a preset range of the target intersection contained in the navigation map information, and combining it with the vehicle kinematics model, a smooth curvature operation is performed on the straight-line trajectory formed by the key points in the multiple target intersections obtained based on the navigation map information, a first turning trajectory for the vehicle to turn can also be generated. Even when the vehicle's own perception and detection are limited, it can still ensure that the vehicle can turn based on the navigation map information, thereby improving the traffic efficiency of the intersection.

[0039] S130: Based on the navigation map information, vehicle positioning information, and vehicle perception information, adjust the first turning trajectory to obtain the second turning trajectory.

[0040] Specifically, according to the description of navigation map information, vehicle positioning information, and vehicle perception information in step S110, by combining vehicle positioning information and navigation map information, the lane information around the vehicle when it reaches the stop line of the target intersection can be obtained. By combining vehicle positioning information and vehicle perception information, the actual situation of the intersection when the vehicle reaches the stop line of the target intersection can be obtained, namely, perception information such as lane line information, curb information, traffic light information, and obstacle information ahead, including obtaining the actual distance information between the vehicle and objects such as lane lines, curbs, traffic lights, and obstacles ahead. Therefore, after obtaining the first turning trajectory based on navigation map information, the first turning trajectory is adjusted by combining navigation map information, vehicle positioning information, and vehicle perception information to obtain a more refined second turning trajectory that better matches the actual driving environment of the vehicle. This improves the robustness of turning at intersections. For various different turning intersections in the city, the appropriate turning trajectory can be generated in real time based on the real-time acquired vehicle perception information and vehicle positioning information, ensuring that the vehicle can automatically and smoothly drive through the intersection.

[0041] As one implementation method, adjusting the first turning trajectory to obtain the second turning trajectory based on navigation map information, vehicle positioning information, and vehicle perception information can be as follows: Combine navigation map information and vehicle positioning information to obtain the vehicle's current turning lane; then, obtain the vehicle's real-time position based on the vehicle positioning information; and then, combine vehicle perception information to confirm whether the vehicle is already in the turning lane and has reached the stop line of the target intersection. When the vehicle reaches the stop line of the target intersection, adjust the first turning trajectory based on the vehicle perception information obtained from the vehicle's position at the stop line of the target intersection to obtain the second turning trajectory. Specifically, taking vehicle perception information including zebra crossings, curbs, obstacles, stop lines, traffic lights, etc. at the target intersection as an example, the first turning trajectory can be adjusted for trajectory avoidance constraints based on curb, zebra crossing, and obstacle information, so that the generated second turning trajectory can bypass or avoid curbs, zebra crossings, and obstacles. The first turning trajectory can also be adjusted for trajectory deceleration or braking constraints based on stop line and traffic light information, so that the generated second turning trajectory can cause the vehicle to decelerate, brake, or proceed at an appropriate location based on traffic light signals.

[0042] Therefore, this application provides a method for planning a turning trajectory at an intersection, applied to a vehicle. The method includes: acquiring navigation map information, vehicle positioning information, and vehicle perception information, wherein the vehicle perception information is used to characterize the environmental information of the target intersection where the vehicle is located, and the target intersection is the intersection where the vehicle is currently located; generating a first turning trajectory corresponding to the target intersection based on the navigation map information, wherein the navigation map information includes at least information on multiple key points within a preset range of the target intersection; and adjusting the first turning trajectory based on the navigation map information, vehicle positioning information, and vehicle perception information to obtain a second turning trajectory. Therefore, this application generates a first turning trajectory for passing through a target intersection using navigation map information. Based on the first turning trajectory, the vehicle positioning information and vehicle perception information are used to adjust the first turning trajectory to obtain a more accurate second turning trajectory that better reflects the actual traffic conditions on the road. This allows the vehicle to adjust and generate appropriate turning trajectories in real time for different intersections, enabling automatic turning at intersections and improving the robustness of turning in autonomous driving. Furthermore, since the navigation map information is obtained based on standard definition navigation maps, automatic turning behavior at intersections can be achieved even without high-definition maps, reducing the cost of autonomous driving.

[0043] Please see Figure 4 , Figure 4 This invention illustrates a flowchart of another intersection turning trajectory planning method provided in an embodiment of this application. The method can be applied to, for example... Figure 1 On the vehicle shown, specifically, the method includes the following steps:

[0044] S210: Obtain navigation map information, vehicle positioning information, and vehicle perception information. The vehicle perception information is used to characterize the environmental information of the surrounding roads and target intersections of the vehicle. The target intersection is the intersection of the road where the vehicle is currently located.

[0045] S220: Based on the navigation map information, generate a first turning trajectory corresponding to the target intersection, wherein the navigation map information includes at least information on multiple key points within a preset range of the target intersection.

[0046] The specific implementation of steps S210-S220 can be referred to the specific implementation of steps S110-S120 in the above embodiments, and will not be repeated here.

[0047] S230: Based on the navigation map information and the vehicle positioning information, determine the current turning lane of the vehicle; based on the vehicle perception information obtained when the vehicle is driving in the turning lane, determine the stop line position of the turning lane.

[0048] As one implementation method, vehicle perception information may include, but is not limited to, perception information of objects such as lane lines, stop lines, zebra crossings, curbs, obstacles, traffic lights, and ground markings around the target intersection, as well as the relative position information of these objects and the vehicle. Furthermore, the navigation map information also includes lane information, reflecting the lane attributes of each lane related to the target intersection, and the vehicle's current position can be determined based on the vehicle's positioning information. Therefore, by combining navigation map information and vehicle positioning information, the current turning lane of the vehicle can be obtained. Based on the vehicle perception information obtained while the vehicle is traveling in the turning lane, the stop line position of the turning lane can be further determined. In other words, by combining vehicle positioning information and vehicle perception information obtained while the vehicle is traveling in the turning lane, the time when the vehicle reaches the stop line position of its current turning lane can be clearly determined.

[0049] S240: When it is determined that the vehicle has reached the stop line position based on the vehicle positioning information, the first turning trajectory is adjusted based on at least one of the vehicle positioning information and the vehicle perception information obtained when the vehicle is driving in the turning lane to obtain the second turning trajectory.

[0050] In this embodiment, based on the above analysis, by combining vehicle positioning information with vehicle perception information acquired when the vehicle is traveling in the turning lane, the exact moment when the vehicle reaches the stop line position of the turning lane can be determined. Therefore, acquiring vehicle positioning and perception information when the vehicle is traveling in the turning lane, and further acquiring vehicle positioning and perception information when the vehicle reaches the stop line position of the turning lane, allows for a more accurate reflection of the vehicle's position within the turning lane and the target intersection, as well as the accurate road conditions of the turning lane and the target intersection. Therefore, when the vehicle reaches the stop line position, adjusting the first turning trajectory based on at least one of the vehicle positioning information and the vehicle perception information acquired while the vehicle is traveling in the turning lane can yield a second turning trajectory that better matches the conditions of the turning lane and the target intersection.

[0051] As one implementation method, when the vehicle reaches the stop line, its current position can be determined based on the vehicle's positioning information. This current position is then used as the starting point for the turn. The first turning trajectory is then adjusted based on this starting point to obtain a second turning trajectory that includes the starting point. Specifically, after determining the vehicle's current position at the stop line of the turning lane, this current position can be used as the starting point. A suitable curve fitting model is used to adjust and fit the first turning trajectory, making its curve close to or even including the starting point. This adjusted curve trajectory is then used as the second turning trajectory, meaning the second turning trajectory includes the starting point. For example, if the first turning trajectory is a simple smooth curve, a multinomial regression model, exponential model, or logarithmic model can be used for fitting. If the first turning trajectory cannot be well fitted to the starting point, a more complex model, such as a neural network model, can be used. Understandably, since the first turning trajectory is generated based on rough key point information in the navigation map when the vehicle is still some distance from the stop line, there may be some error between it and the actual driving position of the vehicle. Therefore, when the vehicle reaches the stop line, the vehicle's current actual position is used as the starting point of the turn to fit and adjust the first turning trajectory, so that the adjusted second turning trajectory includes the starting point of the turn, thereby making the second turning trajectory more consistent with the actual driving situation of the vehicle.

[0052] In one implementation method, when the vehicle reaches the stop line, the position of the curb at the target intersection is determined based on the vehicle perception information acquired while the vehicle is traveling in the turning lane. Then, based on the curb position at the target intersection, the first turning trajectory is adjusted to obtain a second turning trajectory, ensuring that the second turning trajectory does not coincide with the curb position at the target intersection. Specifically, at the moment the vehicle reaches the stop line of the turning lane, the curb position at the target intersection can be determined based on the vehicle perception information acquired while the vehicle is traveling in the turning lane, and more specifically, based on the curb information acquired at the stop line of the turning lane. Then, the curb position at the target intersection is determined based on the curb information in the vehicle perception information. The first turning trajectory is then fitted and adjusted using a curve fitting model, ensuring that the curve of the first turning trajectory does not coincide with the curb position at the target intersection. The adjusted curve trajectory is then used as the second turning trajectory, meaning that the second turning trajectory does not coincide with the curb position at the target intersection. Understandably, since the first turning trajectory is generated based on rough key point information in the navigation map, it means that the actual location of the road edge at the target intersection and its impact on the turning trajectory were not considered during the generation of the first turning trajectory. Therefore, when the vehicle reaches the stop line, the first turning trajectory is fitted and adjusted based on the actual location of the road edge at the target intersection, so that the adjusted second turning trajectory does not coincide with the location of the road edge at the target intersection, thus making the second turning trajectory more consistent with the actual driving situation of the vehicle.

[0053] Therefore, this application proposes an intersection turning trajectory planning method, which is applied to a vehicle. The method includes: acquiring navigation map information, vehicle positioning information, and vehicle perception information, wherein the vehicle perception information is used to characterize the environmental information of the target intersection where the vehicle is located, and the target intersection is the intersection where the vehicle is currently located; generating a first turning trajectory corresponding to the target intersection based on the navigation map information, wherein the navigation map information includes at least information on multiple key points within a preset range of the target intersection; determining the turning lane where the vehicle is currently located based on the navigation map information and the vehicle positioning information; determining the stop line position of the turning lane based on the vehicle perception information acquired while the vehicle is traveling in the turning lane; and adjusting the first turning trajectory to obtain a second turning trajectory based on at least one of the vehicle positioning information and the vehicle perception information acquired while the vehicle is traveling in the turning lane, when the vehicle is determined to have reached the stop line position based on the vehicle positioning information. Therefore, this application generates a first turning trajectory for passing through the target intersection using a low-precision navigation map. When the vehicle reaches the stop line, the first turning trajectory is adjusted based on at least one of the vehicle positioning information and the vehicle perception information obtained when the vehicle is driving in the turning lane to obtain a second turning trajectory. This makes the second turning trajectory more consistent with the turning lane and the target intersection, thereby improving the accuracy of the turning trajectory generated by the low-precision navigation map.

[0054] Please see Figure 5 , Figure 5 This paper illustrates a flowchart of another intersection turning trajectory planning method provided in an embodiment of this application. The method can be applied to, for example... Figure 1 On the vehicle shown, specifically, the method includes the following steps:

[0055] S310: Obtain navigation map information, vehicle positioning information, and vehicle perception information. The vehicle perception information is used to characterize the environmental information of the surrounding roads and target intersections of the vehicle. The target intersection is the intersection of the road where the vehicle is currently located.

[0056] S320: Based on navigation map information, generate a first turning trajectory corresponding to the target intersection, wherein the navigation map information includes information on at least several key points within a preset range of the target intersection.

[0057] S330: Based on the navigation map information, vehicle positioning information, and vehicle perception information, adjust the first turning trajectory to obtain the second turning trajectory.

[0058] The specific implementation of steps S310-S330 can be referred to the specific implementation of steps S110-S130 in the above embodiments, and will not be repeated here.

[0059] S340: Based on the second turning trajectory, control the vehicle to turn at the target intersection, and update the second turning trajectory based on the vehicle perception information obtained during the turning process, so that the vehicle completes the turning behavior according to the updated second turning trajectory.

[0060] In this embodiment, after obtaining the second turning trajectory, the vehicle is controlled to turn at the target intersection based on the second turning trajectory. It is understood that, since unexpected situations may still occur at the target intersection during the vehicle's turning process, such as pedestrians appearing on the zebra crossing or obstacles in the intersection, the second turning trajectory is continuously updated in real time based on the vehicle perception information obtained during the turning process. This ensures that the vehicle continues to turn according to the updated second turning trajectory until the turn is completed, guaranteeing that the turning trajectory can handle various actual road conditions at the target intersection.

[0061] As one implementation method, based on vehicle perception information acquired during the vehicle's turning behavior, the location of obstacles at the target intersection can be determined. Then, the second turning trajectory is updated based on the obstacle locations at the target intersection to ensure that the updated second turning trajectory does not coincide with the obstacle locations at the target intersection. For example, taking vehicle perception information including zebra crossings, obstacles, traffic lights, etc., at the target intersection, the second turning trajectory can be adjusted for trajectory avoidance constraints based on zebra crossing and obstacle information, so that the updated second turning trajectory can bypass or avoid zebra crossings and obstacles. The first turning trajectory can be adjusted for trajectory deceleration or braking constraints based on zebra crossing and traffic light information, so that the generated second turning trajectory can cause the vehicle to decelerate, brake, or proceed at an appropriate location based on zebra crossing or traffic light signals.

[0062] Please see Figure 6 , Figure 6 This paper shows a structural block diagram of a road damage identification device 400 according to an embodiment of this application, including:

[0063] The information acquisition module 401 is used to acquire navigation map information, vehicle positioning information and vehicle perception information. The vehicle perception information is used to characterize the environmental information of the surrounding roads and target intersections of the vehicle. The target intersection is the intersection of the road where the vehicle is currently located.

[0064] The trajectory generation module 402 is used to generate a first turning trajectory corresponding to the target intersection based on navigation map information, wherein the navigation map information includes at least information on multiple key points within a preset range of the target intersection;

[0065] The trajectory adjustment module 403 is used to adjust the first turning trajectory based on the navigation map information, vehicle positioning information and vehicle perception information to obtain the second turning trajectory.

[0066] Optionally, the trajectory generation module 402 is further configured to obtain information on multiple key points within a preset range of the target intersection based on the navigation map information; perform a smoothing operation on the sub-trajectory route between each pair of adjacent key points based on the vehicle kinematics model; and use each sub-trajectory route after the smoothing operation as the first turning trajectory of the target intersection.

[0067] Optionally, the trajectory adjustment module 403 is further configured to determine the current turning lane of the vehicle based on the navigation map information and the vehicle positioning information; determine the stop line position of the turning lane based on the vehicle perception information obtained when the vehicle is driving in the turning lane; and when it is determined that the vehicle has reached the stop line position based on the vehicle positioning information, adjust the first turning trajectory based on at least one of the vehicle positioning information and the vehicle perception information obtained when the vehicle is driving in the turning lane to obtain a second turning trajectory.

[0068] Optionally, the trajectory adjustment module 403 is further configured to determine the current position of the vehicle based on the vehicle positioning information, and use the current position of the vehicle as the turning starting point; adjust the first turning trajectory based on the turning starting point to obtain the second turning trajectory, wherein the second turning trajectory includes the turning starting point.

[0069] Optionally, the trajectory adjustment module 403 is further configured to determine the position of the curb in the target intersection based on the vehicle perception information obtained when the vehicle is driving in the turning lane; and adjust the first turning trajectory based on the position of the curb in the target intersection to obtain the second turning trajectory, wherein the second turning trajectory does not coincide with the position of the curb in the target intersection.

[0070] Optionally, the trajectory adjustment module 403 is further configured to control the vehicle to make a turning behavior at the target intersection according to the second turning trajectory, and update the second turning trajectory based on the vehicle perception information obtained during the turning behavior, so that the vehicle completes the turning behavior according to the updated second turning trajectory.

[0071] Optionally, the trajectory adjustment module 403 is further configured to determine the position of the obstacle in the target intersection based on the vehicle perception information obtained during the vehicle's turning behavior; and update the second turning trajectory based on the position of the obstacle in the target intersection so that the updated second turning trajectory does not coincide with the position of the obstacle in the target intersection.

[0072] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the above-described device and module can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0073] In the several embodiments provided in this application, the coupling between modules can be electrical, mechanical, or other forms of coupling.

[0074] Furthermore, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module. The integrated modules described above can be implemented in hardware or as software functional modules.

[0075] Please see Figure 7 This diagram illustrates a structural block diagram of a computer-readable storage medium provided in an embodiment of this application. The computer-readable medium 500 stores program code that can be called by a processor to execute the methods described in the above method embodiments.

[0076] The computer-readable storage medium 500 may be an electronic memory such as flash memory, EEPROM (Electrically Erasable Programmable Read-Only Memory), EPROM, hard disk, or ROM. Optionally, the computer-readable storage medium 500 includes a non-transitory computer-readable storage medium. The computer-readable storage medium 500 has storage space for program code 510 that performs any of the method steps described above. This program code can be read from or written to one or more computer program products. The program code 510 may be compressed, for example, in a suitable form.

[0077] As one implementation method, this application also provides a computer program product. This computer program product includes a computer program / instructions that, when executed by a processor, implement the steps of the above-described method.

[0078] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A method for intersection turn trajectory planning, the method comprising: The method is applied to a vehicle and comprises the following steps: Obtaining navigation map information, vehicle positioning information and vehicle perception information, wherein the vehicle perception information is used to represent environmental information of a target intersection where the vehicle is located, and the target intersection is a turning intersection of a road where the vehicle is currently located; Based on the navigation map information, a first turning trajectory corresponding to the target intersection is generated, wherein the navigation map information at least includes information of a plurality of key points within a preset range of the target intersection; Based on the navigation map information, the vehicle positioning information and the vehicle perception information, the first turning trajectory is adjusted to obtain a second turning trajectory; According to the second turning trajectory, the vehicle is controlled to perform a turning behavior at the target intersection, and based on vehicle perception information obtained during the turning behavior of the vehicle, a position of an obstacle in the target intersection is determined; Based on the position of the obstacle in the target intersection, the second turning trajectory is updated so that the vehicle completes the turning behavior according to the updated second turning trajectory; wherein the updated second turning trajectory does not coincide with the position of the obstacle in the target intersection.

2. The method of claim 1, wherein, The method comprises the following steps: According to the navigation map information, information of a plurality of key points within a preset range of the target intersection is obtained; Based on a vehicle kinematics model, a smoothing operation is performed on a sub-trajectory route between each adjacent two key points; After each sub-trajectory route is subjected to the smoothing operation, the sub-trajectory route is taken as the first turning trajectory of the target intersection.

3. The method of claim 1, wherein, The method comprises the following steps: Based on the navigation map information and the vehicle positioning information, a turning lane where the vehicle is currently located is determined; Based on vehicle perception information obtained when the vehicle is driving in the turning lane, a stop line position of the turning lane is determined; When it is determined according to the vehicle positioning information that the vehicle drives to the stop line position, at least one of the vehicle positioning information and the vehicle perception information obtained when the vehicle is driving in the turning lane is used to adjust the first turning trajectory to obtain a second turning trajectory.

4. The method of claim 3, wherein, The method comprises the following steps: Based on the vehicle positioning information, a current position of the vehicle is determined, and the current position of the vehicle is taken as a turning starting point; Based on the turning starting point, the first turning trajectory is adjusted to obtain the second turning trajectory, and the second turning trajectory includes the turning starting point.

5. The method of claim 3, wherein, The method comprises the following steps: Based on the vehicle perception information obtained when the vehicle is driving in the turning lane, a road edge position in the target intersection is determined; Based on the road edge position in the target intersection, the first turning trajectory is adjusted to obtain the second turning trajectory, and the second turning trajectory does not coincide with the road edge position in the target intersection.

6. An intersection turn trajectory planning apparatus characterized by comprising: The device is applied to a vehicle and comprises: An information acquisition module is configured to acquire navigation map information, vehicle positioning information, and vehicle perception information, wherein the vehicle perception information is used to represent environmental information of a target intersection where the vehicle is located, and the target intersection is a turning intersection of a road where the vehicle is currently located; A trajectory generation module is configured to generate a first turning trajectory corresponding to the target intersection based on the navigation map information, wherein the navigation map information at least includes information of a plurality of key points within a preset range of the target intersection; A trajectory adjustment module is configured to adjust the first turning trajectory based on the navigation map information, the vehicle positioning information, and the vehicle perception information to obtain a second turning trajectory; control the vehicle to perform a turning behavior at the target intersection according to the second turning trajectory; determine an obstacle position in the target intersection according to vehicle perception information acquired during the turning behavior of the vehicle; and update the second turning trajectory based on the obstacle position in the target intersection, so that the vehicle completes the turning behavior according to the updated second turning trajectory; and the updated second turning trajectory does not coincide with the obstacle position in the target intersection.

7. A vehicle characterized by comprising: The device comprises: a perception sensor configured to acquire vehicle perception information; one or more processors; a memory; one or more application programs, wherein the one or more application programs are stored in the memory and configured to be executed by the one or more processors, and the one or more application programs are configured to perform the method according to any one of claims 1-5.

8. A computer-readable storage medium, characterized in that, The computer readable storage medium stores processor-executable program code, and the program code is executed by the processor to make the processor perform the method according to any one of claims 1-5.

Images