Automatic driving based turning trajectory generation method and device, equipment and medium

Abstract

The application discloses a method, device and equipment for generating a U-turn trajectory based on automatic driving and a medium. The method comprises the following steps: when a vehicle drives to a preset U-turn section, a U-turn space is discretely sampled to obtain a sampling point, wherein the U-turn space is a space between a current lane of the vehicle and an opposite lane in which the vehicle can make a U-turn, and the sampling point is a position on a center line of the U-turn space that is passed by the vehicle when making a U-turn; a vehicle adjustment space is constructed according to a total width of the current lane of the vehicle and the opposite lane and a length of the U-turn space; and a vehicle U-turn trajectory is generated by a preset path planning method according to a current position of the vehicle, the sampling point, the vehicle adjustment space and a road condition of a lane to be driven into. The method can solve the problem that a U-turn trajectory cannot be flexibly generated in the prior art.

Description

Technical Field

[0001] This invention relates to the field of autonomous driving technology, and in particular to a method, apparatus, device, and medium for generating U-turn trajectories based on autonomous driving. Background Technology

[0002] With the rapid advancement of technology, autonomous driving has emerged, and the generation of U-turn trajectories is one of its core functions. However, in actual U-turns, the turning point is often filled with various obstacles, such as illegally parked vehicles and street vendors. These factors often lead to complex road conditions such as narrow roads and congestion. Current technologies, however, often rely solely on the static environment of the road structure when generating U-turn trajectories. This results in the generated U-turn trajectory often failing to complete the turn successfully in one go under complex and varied road conditions, and may even lead to the vehicle getting stuck at the turning point, causing traffic congestion and safety hazards. This not only affects vehicle traffic efficiency but also places additional pressure on the entire transportation system. Summary of the Invention

[0003] This invention provides a method, apparatus, device, and medium for generating U-turn trajectories based on autonomous driving, aiming to solve the problem of the inability to flexibly generate U-turn trajectories in the prior art.

[0004] In a first aspect, embodiments of the present invention provide a method for generating a U-turn trajectory based on autonomous driving. The method is applied to an autonomous driving system of a vehicle and includes: when the vehicle travels to a preset U-turn section, discretely sampling the U-turn space to obtain sampling points, wherein the U-turn space is the space between the vehicle's current lane and the oncoming lane where a U-turn can be made, and the sampling points are the positions that the vehicle passes through when making a U-turn on the center line of the U-turn space; constructing a vehicle adjustment space based on the sum of the widths of the vehicle's current lane and the oncoming lane and the length of the U-turn space; and generating a vehicle U-turn trajectory based on the vehicle's current position, the sampling points, the vehicle adjustment space, and the road conditions of the lane to be entered using a preset path planning method.

[0005] Secondly, embodiments of the present invention also provide a U-turn trajectory generation device based on autonomous driving, comprising: a sampling unit, used to discretely sample the U-turn space to obtain sampling points when the vehicle travels to a preset U-turn section, wherein the U-turn space is the space between the vehicle's current lane and the opposite lane where a U-turn can be made, and the sampling points are the positions that the vehicle must pass through when making a U-turn on the center line of the U-turn space; a construction unit, used to construct a vehicle adjustment space based on the sum of the widths of the vehicle's current lane and the opposite lane and the length of the U-turn space; and a planning unit, used to generate a vehicle U-turn trajectory based on the vehicle's current position, the sampling points and the vehicle adjustment space, and the road conditions of the lane to be entered, using a preset path planning method.

[0006] Thirdly, embodiments of the present invention also provide a computer device, which includes a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the above-described method.

[0007] Fourthly, embodiments of the present invention also provide a computer-readable storage medium storing a computer program, the computer program including program instructions that, when executed by a processor, can implement the above-described method.

[0008] This invention provides a method, apparatus, device, and medium for generating U-turn trajectories based on autonomous driving. The method is applied to an autonomous driving system for vehicles and includes: when the vehicle travels to a preset U-turn section, discretely sampling the U-turn space to obtain sampling points, wherein the U-turn space is the space between the vehicle's current lane and the oncoming lane where a U-turn is possible, and the sampling points are the positions the vehicle passes through when making a U-turn on the center line of the U-turn space; constructing a vehicle adjustment space based on the sum of the widths of the vehicle's current lane and the oncoming lane and the length of the U-turn space; and generating a vehicle U-turn trajectory using a preset path planning method based on the vehicle's current position, the sampling points, the vehicle adjustment space, and the road conditions of the lane to be entered. This invention, through sampling of U-turn spaces, obtains sampling points to facilitate the generation of U-turn trajectories with different curvatures based on these sampling points. It also constructs a vehicle adjustment space limit to restrict the area occupied by vehicle adjustments, thus avoiding traffic congestion. Based on the road conditions of the lane to be entered and information such as sampling points, a preset path planning method generates the vehicle U-turn trajectory. This allows for flexible generation of U-turn trajectories that can be used for vehicle U-turns according to different road conditions, significantly improving the efficiency of vehicle U-turns and avoiding traffic congestion caused by unsuccessful U-turns. Attached Figure Description

[0009] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0010] Figure 1 This is a flowchart illustrating the method for generating a U-turn trajectory based on autonomous driving, as provided in an embodiment of the present invention.

[0011] Figure 2 This is a schematic diagram of a sub-process of the method for generating U-turn trajectories based on autonomous driving provided in an embodiment of the present invention;

[0012] Figure 3 This is a schematic diagram of a sub-process of the method for generating U-turn trajectories based on autonomous driving provided in an embodiment of the present invention;

[0013] Figure 4 This is a schematic diagram of a sub-process of the method for generating U-turn trajectories based on autonomous driving provided in an embodiment of the present invention;

[0014] Figure 5 This is a schematic diagram of a sub-process of the method for generating U-turn trajectories based on autonomous driving provided in an embodiment of the present invention;

[0015] Figure 6 This is a schematic diagram of a sub-process of the method for generating U-turn trajectories based on autonomous driving provided in an embodiment of the present invention;

[0016] Figure 7 This is a schematic diagram of a sub-process of the method for generating U-turn trajectories based on autonomous driving provided in an embodiment of the present invention;

[0017] Figure 8 A schematic block diagram of a U-turn trajectory generation device based on autonomous driving provided in an embodiment of the present invention;

[0018] Figure 9 A schematic block diagram of a computer device provided for an embodiment of the present invention. Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] It should be understood that, when used in this specification and the appended claims, the terms "comprising" and "including" indicate the presence of the described features, integrals, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or collections thereof.

[0021] It should also be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0022] It should also be further understood that the term "and / or" as used in this specification and the appended claims refers to any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0023] Please see Figure 1 , Figure 1 This is a flowchart illustrating the U-turn trajectory generation method based on autonomous driving provided in this embodiment of the invention. The U-turn trajectory generation method based on autonomous driving in this embodiment can be applied to vehicles equipped with autonomous driving technology. If traditional methods are used to generate U-turn paths, it may lead to problems such as the inability to successfully make a U-turn in one attempt, affecting traffic. This method can intelligently plan the optimal U-turn trajectory, enabling autonomous vehicles to complete the U-turn smoothly and safely, thereby effectively improving road traffic efficiency and safety.

[0024] Figure 1 This is a flowchart illustrating the method for generating a U-turn trajectory based on autonomous driving provided in an embodiment of the present invention. As shown in the figure, the method includes the following steps S110-S130.

[0025] S110. When a vehicle travels to a preset U-turn section, the U-turn space is discretely sampled to obtain sampling points. The U-turn space is the space between the vehicle's current lane and the opposite lane where a U-turn can be made. The sampling points are the positions that the vehicle passes through when making a U-turn on the center line of the U-turn space.

[0026] In this embodiment, the preset U-turn section is a location where a vehicle can make a U-turn, as identified by the semantic map integrated into the vehicle's autonomous driving system. The semantic map contains rich road information, such as lane centerlines, roadbeds, and road markings. The discrete sampling is the process of acquiring a number of sampling points at preset discrete intervals. It is understood that in road sections where U-turns are permitted, a space is typically defined between the current lane and the oncoming lane; this space is the U-turn space. Sampling points are acquired at the centerline of this U-turn space or at the boundary between the current lane and the oncoming lane, at preset discrete intervals. These sampling points represent the locations where the vehicle passes through the centerline or boundary line during the U-turn. Acquiring sampling points allows for a more accurate and comprehensive depiction of the vehicle's path trajectory with different curvatures during the U-turn, thereby ensuring the smoothness and safety of the U-turn process.

[0027] S120. Construct a vehicle adjustment space based on the sum of the widths of the lane where the vehicle is currently located and the opposite lane, as well as the length of the U-turn space.

[0028] In this embodiment, the vehicle adjustment space is the space within which the vehicle can freely reverse and maneuver at a U-turn point, without strictly adhering to the road network. The primary purpose is to allow the vehicle to adjust its posture within this free space for a U-turn. The vehicle adjustment space is rectangular, with its length equal to the length of the U-turn space and its width equal to the sum of the widths of the vehicle's current lane and the oncoming lane. It can be understood that the vehicle adjustment space is also divided into many smaller spaces, with each lane potentially constituting a smaller space. The goal is to move the vehicle within the smallest possible area when adjusting its posture to avoid obstructing traffic in the normal lanes. By constructing the vehicle adjustment space based on the U-turn space and the lane width, the space for the vehicle to adjust its posture is limited, preventing the vehicle from occupying too much space and causing traffic congestion or failed U-turns.

[0029] S130. Based on the vehicle's current position, the sampling point and the vehicle's adjustment space, and the road conditions of the lane to be entered, a vehicle turning trajectory is generated using a preset path planning method.

[0030] In this embodiment, the preset path planning method is an algorithm that calculates the path from one posture to another for a vehicle under a fixed turning radius, such as hybridA*, RS curve, and other path planning methods. Specifically, based on the current vehicle position, the sampling point, and the road conditions of the lane to be entered, the hybridA* algorithm generates several vehicle turning trajectories to reach the lane to be entered. The hybridA* algorithm combines the efficient search characteristics of the A* algorithm with the flexibility of dynamic programming, and can take into account the vehicle's dynamic constraints (such as maximum steering angle, acceleration limits, etc.) in complex environments, thereby generating trajectories that better match actual driving conditions. Among the generated vehicle turning trajectories, a trajectory that allows for vehicle posture adjustment within the vehicle's adjustment space is selected as a backup vehicle turning trajectory, ensuring that the vehicle does not exceed its physical limitations or conflict with other road users when performing a turning maneuver. Based on the road conditions of the lane to be entered, the shortest path or the shortest route among the backup vehicle turning trajectories is selected as the target turning trajectory, so that the vehicle performs a turning maneuver according to the target turning trajectory. By generating a vehicle U-turn trajectory based on the vehicle's current position, the sampling point, the vehicle's adjustment space, the road conditions of the lane to be entered, and a preset path planning method, the accurate tracking of the trajectory and the smooth completion of the U-turn are ensured.

[0031] In one embodiment, such as Figure 2 As shown, step S130 further includes steps S131-S132.

[0032] S131. If there is an obstacle in the lane to be entered, the volume of the obstacle is obtained by volume prediction based on the center coordinates of the obstacle and the projection of the obstacle.

[0033] S132. Based on the vehicle's current position, the volume of the obstacle, the sampling point, and the vehicle's adjustment space, a vehicle turning trajectory around the obstacle is generated using the preset path planning method.

[0034] In this embodiment, the obstacle is an object that affects the vehicle's driving or U-turn, such as illegally parked vehicles or street vendors. The obstacle can be identified by the vehicle's onboard LiDAR and camera. The autonomous driving system will perceive the real-time position of the obstacle in the environment, and then obtain the volume of the obstacle based on the perceived projection and coordinates. The specific method for obtaining the volume is not limited, as long as the volume of the obstacle can be obtained. Based on the vehicle's current position, the volume of the obstacle, the sampling point, and the vehicle's adjustment space, a vehicle U-turn trajectory is generated using the preset path planning method. Specifically, a shortest trajectory can be planned at each sampling point using methods such as hybridA* combined with vehicle kinematics to connect to the oncoming lane. If an obstacle causes the curvature to be insufficient, it is filtered out, and the failure to successfully make a U-turn at that sampling point is recorded. If a U-turn can be made directly to connect to the oncoming lane, the successfully planned trajectory is returned, thus generating a vehicle U-turn trajectory that bypasses the obstacle. By using information such as the vehicle's current position, the size of the obstacle, and the preset path planning method, a U-turn trajectory can be generated to bypass the obstacle, thereby helping the vehicle to successfully make a U-turn and achieving the purpose of flexibly planning a unique U-turn trajectory.

[0035] In one embodiment, such as Figure 3 As shown, step S130 further includes steps S133-S136.

[0036] S133. Determine whether the lane to be entered is a congested lane based on the number of vehicles and their speed in the lane to be entered.

[0037] S134. If it is a congested lane, then the lane that is in the same direction as the lane to be entered and is not congested will be identified as the new lane to be entered.

[0038] S135. Determine the lane entry position in the new lane to be entered based on the vehicle's current position and preset attitude angle;

[0039] S136. Generate the vehicle U-turn trajectory using the preset path planning method based on the vehicle's current position, lane entry position, sampling point, and vehicle adjustment space.

[0040] In this embodiment, the number of vehicles is the total number of vehicles in each lane within a certain range of the lane to be entered, and the driving speed is the speed of each vehicle. The system determines whether the lane to be entered is a congested lane based on the number of vehicles and their speeds. Specifically, the traffic density within a preset driving range is obtained by subtracting the sum of the lengths of all vehicles within the preset driving range from the distance between the first and last vehicles. The vehicle density is then obtained by dividing the difference by the total number of vehicles. The system determines whether the vehicle density and the driving speed are significantly less than a preset vehicle density and a preset average driving speed. If both are less than these values, the lane to be entered is determined to be a congested lane. If the lane to be entered is a congested lane, then a lane traveling in the same direction as the lane to be entered that is not congested will be designated as the new lane to be entered. For example, if the third lane from the left is the lane to be entered, but the second and / or third lane from the left are congested lanes, then the first lane from the left, traveling in the same direction as the third lane from the left, will be a non-congested lane, or its congestion level will be lower than that of the third and second lanes from the left. In this case, the first lane from the left can be designated as the new lane to be entered. The lane entry position is determined in the new lane to be entered based on the vehicle's current position and a preset attitude angle. For example, the preset attitude angle can be 40 degrees with the direction of the opposite lane. The vehicle U-turn trajectory is generated using the preset path planning method based on the vehicle's current position, lane entry position, sampling points, and vehicle adjustment space. Specifically, U-turn trajectories can be planned from the vehicle's current position using trajectory planning methods such as hybrid A* and RS curves, passing through various sampling points to reach the lane entry position. Then, the optimal vehicle trajectory is selected from these U-turn trajectories according to preset selection criteria. For example, the trajectory with the shortest travel distance or the fewest gear shifts can be selected as the optimal vehicle trajectory. By selecting a non-congested lane in the same direction as the lane to be entered as the new lane when the lane to be entered is congested, the vehicle can flexibly cope with traffic congestion. The preset path planning method generates a U-turn trajectory that allows the vehicle to maneuver and maneuver within the vehicle adjustment space, enabling the vehicle to make a U-turn flexibly and smoothly to avoid affecting traffic.

[0041] In one embodiment, such as Figure 4 As shown, step S133 is followed by steps S1331-S1333.

[0042] S1331. If the lane to be entered is a congested lane, then the lane to be entered is subjected to vertical projection based on the length of the U-turn space to obtain a search point.

[0043] S1332. Search for other vehicles ahead of the lane to be entered based on the search point and record the search distance. If the search distance is greater than the preset search distance, determine the queuing point based on the last vehicle in the lane to be entered.

[0044] S1333. Generate the vehicle turning trajectory using the preset path planning method based on the vehicle's current location, the queuing point, the sampling point, and the vehicle's adjustment space.

[0045] In this embodiment, the search point is the starting point for searching for vehicles. Specifically, when the lane to be entered is a congested lane, a search point can be obtained by projecting a perpendicular line onto the lane to be entered based on the length of the U-turn space. Specifically, a perpendicular line can be drawn from one end of the U-turn space to the lane line of the lane to be entered. This perpendicular line intersects the lane line at a point, which is the search point. Vehicles are then searched for ahead of the lane to be entered based on this search point until the first vehicle is found, and the search distance is returned. If the search distance is greater than a preset search distance, it is considered that although the lane to be entered is congested, there is still a certain distance between vehicles. In this case, a queuing point can be found behind the last vehicle in the congested lane. The preset search distance is a pre-set distance that allows for queuing. The vehicle's U-turn trajectory is generated using the preset path planning method based on the vehicle's current position, the queuing point, the sampling point, and the vehicle's adjustment space. Specifically, a suitable trajectory is planned according to the vehicle's current position and speed to reach the queuing point. The shortest trajectory can be selected from the generated U-turn trajectories as the target trajectory, and the vehicle is controlled to proceed to the last queuing point in the congestion queue to make a U-turn. By obtaining the queuing point based on the search point, it is determined whether the lane to be entered is only a congested area. If it is a congested area, the vehicle detours to the queuing point after the congestion to join the queue, avoiding forcibly cutting in with U-turning vehicles at the current position. By generating the U-turn trajectory based on the queuing point, the vehicle U-turn can be flexibly implemented.

[0046] In one embodiment, such as Figure 5 As shown, step S130 further includes steps S137-S138.

[0047] S137. If the current driving lane is a congested lane, identify the vehicles waiting to make a U-turn by comparing the position and orientation of the vehicles in front.

[0048] S138. Predict the predicted turning trajectory of the vehicle to be turned around using a preset trajectory prediction method, and turn the vehicle around according to the predicted turning trajectory.

[0049] In this embodiment, the step of determining whether the current driving lane is a congested lane specifically involves the vehicle's autonomous driving system determining whether the current lane is congested based on the number of vehicles and their speed in the lane. If the current driving lane is congested, the vehicle's adjustment space and U-turn space are compressed, making it impossible to freely reverse and adjust its posture within the space. Therefore, it is necessary to identify vehicles waiting to make a U-turn by comparing their positions and orientations with those of vehicles not making U-turns. For example, the orientation of the vehicle in front can be compared with that of vehicles not making U-turns. If the orientation of the vehicle in front is different from that of the vehicles not making U-turns and is biased towards the U-turn space, then the vehicle in front can be identified as the vehicle waiting to make a U-turn. The vehicles waiting to make U-turns can form a U-turn queue, and then several U-turn trajectories of the queue are generated using preset path planning methods such as RS curves. The vehicle can select the optimal U-turn trajectory for the main vehicle from the U-turn trajectories of the U-turn queue to make a U-turn. For example, an optimal trajectory can be selected based on the condition of minimizing the expected number of gear shifts to make a U-turn. By identifying vehicles waiting to make a U-turn in the current lane when there is congestion, and making a U-turn based on the predicted U-turn trajectory of the vehicles waiting to make a U-turn, a U-turn can be achieved without causing traffic congestion.

[0050] In one embodiment, such as Figure 6 As shown, step S130 further includes steps S139-S1310.

[0051] S139. If the vehicle turning trajectory cannot be generated based on the sampling points, the driving range that can enter the lane to be entered is expanded, and the sampling point density is reduced to obtain the latest sampling points.

[0052] S1310. The latest sampling point and the driving range are used to perform trajectory planning through the preset path planning method until the vehicle turning trajectory is generated.

[0053] In this embodiment, the sampling point density refers to the density of the sampling points. Discrete sampling is performed at the centerline of the U-turn space to obtain the sampling points, and the vehicle U-turn trajectory is generated based on the sampling points and other information. If the vehicle U-turn trajectory cannot be generated, it means that a direct U-turn cannot be completed, and the vehicle needs to adjust its posture by reversing or other actions to overcome the vehicle's potential predicament of being stuck in the U-turn space and complete the U-turn. In this case, it is necessary to expand the search range to find a position that can enter the lane to be entered, and generate a trajectory that can be reversed using a preset path planning method such as hybridA*. Furthermore, when a direct U-turn cannot be completed, it is not necessary to obtain sampling points densely to find a vehicle U-turn trajectory that can be directly performed. Therefore, the density of subsequent sampling points is reduced to resample and obtain the latest sampling points, and a relatively reasonable trajectory is planned to the opposite lane based on the latest sampling points and the driving range. The vehicle then reverses within the vehicle adjustment space to complete the U-turn. When a vehicle turning trajectory cannot be generated, trajectory planning is performed by expanding the driving range and reducing the sampling point density until the vehicle turning trajectory is generated. This allows for the generation of a trajectory that allows the vehicle to turn around through operations such as reversing when a direct vehicle turning trajectory cannot be generated, thus helping the vehicle to achieve a U-turn.

[0054] In one embodiment, such as Figure 7 As shown, step S130 is followed by steps S1301-S1302.

[0055] S1301. If it is detected that other vehicles are making U-turns within the U-turn space, the U-turn trajectory of the other vehicles is predicted by a preset neural network to obtain the predicted U-turn path of the other vehicles.

[0056] S1302. Delete the trajectories of several vehicle U-turn trajectories that intersect with the predicted U-turn path.

[0057] In this embodiment, the preset neural network is a neural network combining map networks and vehicle kinematics, such as convolutional neural networks or recurrent neural networks. This is not limited to any particular type, as long as it can predict the turning paths of other vehicles making U-turns. If the radar and camera of the autonomous driving system detect other vehicles making U-turns within the designated space, the main vehicle's U-turn trajectory must avoid these trajectories to prevent traffic accidents. Therefore, after generating the vehicle U-turn trajectory, the preset neural network predicts the U-turn trajectories of other vehicles to obtain their predicted U-turn paths. Specifically, the neural network can predict the predicted U-turn paths of other vehicles based on their attitude angles and positions, and then project these predicted paths onto the U-turn location. Trajectories that intersect with the predicted U-turn paths from the generated vehicle U-turn trajectories are deleted to avoid the movement paths of other vehicles. A suitable path can then be found among the remaining trajectories for the vehicle U-turn. For example, the optimal trajectory can be found among the remaining vehicle U-turn trajectories based on conditions such as shortest travel distance, shortest completion time, and fewest gear shifts. By using a preset neural network to predict the U-turn paths of other vehicles and deleting trajectories that intersect with the predicted U-turn paths, collisions with other vehicles making U-turns can be avoided, thus effectively preventing traffic accidents.

[0058] In one embodiment, after generating the vehicle U-turn trajectory using a preset path planning method based on the vehicle's current position, the sampling point, the vehicle's adjustment space, and the road conditions of the lane to be entered, the method further includes:

[0059] The estimated turning time of the vehicle's turning trajectory is generated based on the lane congestion situation, and a global path is generated and the corresponding global turning time is obtained through the preset global planning method.

[0060] The estimated U-turn time is compared with the global U-turn time. If the estimated U-turn time is less than the global U-turn time, the vehicle makes a U-turn according to the global path.

[0061] In this embodiment, the preset global planning method is a static planning algorithm that performs path planning based on existing map information, finding an optimal path from the starting point to the target point under the maximum curvature that meets the vehicle's kinematic requirements. If the estimated U-turn time is less than the global U-turn time, it indicates that the time consumed by the generated vehicle U-turn trajectory is too long and far less than the time consumed by rerouting. In this case, the vehicle will rerout according to the global path to achieve the U-turn.

[0062] Furthermore, this embodiment also includes planning a new U-turn trajectory based on the vehicle's current position when the vehicle is making a U-turn by reversing. This new U-turn trajectory is generated based on the movement of other vehicles within the vehicle's current adjustment space. If a vehicle blocking the vehicle in front of it starts moving, a new U-turn trajectory can be generated based on the remaining space. The efficiency of the currently selected reversing trajectory is continuously compared with the new U-turn trajectory. If the new U-turn trajectory is more efficient, there is no need to continue reversing; instead, the vehicle can make a U-turn based on the new trajectory. By continuously comparing the efficiency of the vehicle's U-turn trajectory with other U-turn paths, the system helps the vehicle select a more efficient path for the U-turn, thus solving the problem of the inability to flexibly generate U-turn trajectories in existing technologies.

[0063] Figure 8 This is a schematic block diagram of a U-turn trajectory generation device 200 based on autonomous driving provided in an embodiment of the present invention. Figure 8 As shown, corresponding to the above-described method for generating U-turn trajectories based on autonomous driving, the present invention also provides an apparatus for generating U-turn trajectories based on autonomous driving. This apparatus includes a unit for executing the above-described method for generating U-turn trajectories based on autonomous driving, and can be configured in a desktop computer, tablet computer, laptop computer, or other terminal. For details, please refer to... Figure 8 The U-turn trajectory generation device based on autonomous driving includes a sampling unit 210, a construction unit 220, and a planning unit 230.

[0064] The sampling unit 210 is used to perform discrete sampling of the U-turn space to obtain sampling points when the vehicle travels to the preset U-turn section. The U-turn space is the space between the lane where the vehicle is currently located and the opposite lane where a U-turn can be made. The sampling point is the position that the vehicle must pass through when making a U-turn on the center line of the U-turn space.

[0065] The construction unit 220 is used to construct a vehicle adjustment space based on the sum of the widths of the lane where the vehicle is currently located and the opposite lane, as well as the length of the U-turn space.

[0066] Planning unit 230 is used to generate a vehicle U-turn trajectory based on the vehicle's current position, the sampling point and the vehicle's adjustment space, and the road conditions of the lane to be entered, using a preset path planning method.

[0067] In one embodiment, the planning unit 230 includes a volume prediction unit and a first generation subunit.

[0068] A volume prediction unit is used to predict the volume of an obstacle based on the center coordinates of the obstacle and its projection if there is an obstacle in the lane to be entered.

[0069] The first generation subunit is used to generate a vehicle U-turn trajectory around the obstacle based on the vehicle's current position, the volume of the obstacle, the sampling point, and the vehicle's adjustment space using the preset path planning method.

[0070] In one embodiment, the planning unit 230 includes a judgment unit, a lane determination unit, a location determination unit, and a second generation subunit.

[0071] The judgment unit is used to determine whether the lane to be entered is a congested lane based on the number of vehicles and their speed in the lane to be entered.

[0072] The lane determination unit is used to determine a non-congested lane that is in the same direction as the lane to be entered as the new lane if the lane is congested.

[0073] A position determination unit is used to determine a lane entry position in the new lane to be entered based on the vehicle's current position and a preset attitude angle.

[0074] The second generation subunit is used to generate the vehicle U-turn trajectory based on the vehicle's current position, the lane entry position, the sampling point, and the vehicle adjustment space using the preset path planning method.

[0075] In one embodiment, the planning unit 230 includes a search unit, a queuing unit, and a third generation subunit.

[0076] The search unit is used to perform vertical projection processing on the lane to be entered based on the length of the U-turn space to obtain a search point if the lane to be entered is a congested lane.

[0077] The queuing unit is used to search for other vehicles ahead of the lane to be entered based on the search point and record the search distance. If the search distance is greater than the preset search distance, the queuing point is determined based on the last vehicle in the lane to be entered.

[0078] The third generation subunit is used to generate the vehicle turning trajectory based on the vehicle's current position, the queuing point, the sampling point, and the vehicle adjustment space using the preset path planning method.

[0079] In one embodiment, the planning unit 230 includes a U-turn prediction unit and a U-turn unit.

[0080] The U-turn prediction unit is used to identify vehicles that need to make a U-turn by comparing the position and orientation of vehicles ahead if the current driving lane is a congested lane.

[0081] The U-turn unit is used to predict the predicted U-turn trajectory of the vehicle to be turned around using a preset trajectory prediction method, and to turn the vehicle around according to the predicted U-turn trajectory.

[0082] In one embodiment, the planning unit 230 includes an acquisition unit and a planning unit.

[0083] The acquisition unit is configured to, if the vehicle turning trajectory cannot be generated based on the sampling points, expand the driving range that can enter the lane to be entered and reduce the sampling point density to obtain the latest sampling points.

[0084] The planning unit is used to perform trajectory planning on the latest sampling point and the driving range using the preset path planning method until the vehicle turns around trajectory is generated.

[0085] In one embodiment, the planning unit 230 includes a trajectory prediction unit and a deletion unit.

[0086] The trajectory prediction unit is used to predict the turning trajectory of other vehicles by using a preset neural network if it is detected that other vehicles are making a U-turn in the U-turn space.

[0087] The deletion unit is used to delete trajectories from the vehicle U-turn trajectories that intersect with the predicted U-turn path.

[0088] It should be noted that those skilled in the art can clearly understand that the specific implementation process of the above-mentioned autonomous driving-based U-turn trajectory generation device 200 and each unit can be referred to the corresponding description in the foregoing method embodiments. For the sake of convenience and brevity, it will not be repeated here.

[0089] The aforementioned U-turn trajectory generation device based on autonomous driving can be implemented as a computer program, which can, for example... Figure 9 It runs on the computer device shown.

[0090] Please see Figure 9 , Figure 9 This is a schematic block diagram of a computer device provided in an embodiment of this application. The computer device 500 can be a terminal or a server. The terminal can be an electronic device with communication functions, such as a smartphone, tablet, laptop, desktop computer, personal digital assistant, or wearable device. The server can be a standalone server or a server cluster composed of multiple servers.

[0091] See Figure 9 The computer device 500 includes a processor 502, a memory, and a network interface 505 connected via a system bus 501. The memory may include a non-volatile storage medium 503 and internal memory 504.

[0092] The non-volatile storage medium 503 may store an operating system 5031 and a computer program 5032. The computer program 5032 includes program instructions that, when executed, cause the processor 502 to perform a method for generating a U-turn trajectory based on autonomous driving.

[0093] The processor 502 provides computing and control capabilities to support the operation of the entire computer device 500.

[0094] The internal memory 504 provides an environment for the execution of the computer program 5032 in the non-volatile storage medium 503. When the computer program 5032 is executed by the processor 502, the processor 502 can execute a method for generating a U-turn trajectory based on autonomous driving.

[0095] This network interface 505 is used for network communication with other devices. Those skilled in the art will understand that... Figure 9 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device 500 to which the present application is applied. The specific computer device 500 may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0096] The processor 502 is used to run a computer program 5032 stored in a memory to implement the steps of the above method.

[0097] It should be understood that in the embodiments of this application, the processor 502 may be a central processing unit (CPU), or it may be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor.

[0098] It will be understood by those skilled in the art that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program includes program instructions and can be stored in a storage medium, which is a computer-readable storage medium. The program instructions are executed by at least one processor in the computer system to implement the process steps of the embodiments of the above methods.

[0099] Therefore, the present invention also provides a storage medium. This storage medium can be a computer-readable storage medium. The storage medium stores a computer program, wherein the computer program includes program instructions. When executed by a processor, the program instructions cause the processor to perform the steps of the method described above.

[0100] The storage medium can be any computer-readable storage medium capable of storing program code, such as a USB flash drive, portable hard drive, read-only memory (ROM), magnetic disk, or optical disk.

[0101] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0102] In the several embodiments provided by this invention, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative. For example, the division of each unit is merely a logical functional division, and there may be other division methods in actual implementation. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed.

[0103] The steps in the method of this invention can be adjusted, merged, or reduced in order according to actual needs. The units in the device of this invention can be merged, divided, or reduced according to actual needs. Furthermore, the functional units in the various embodiments of this invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0104] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, a terminal, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention.

[0105] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present invention, and these modifications or substitutions should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for generating U-turn trajectories based on autonomous driving, characterized in that, The method includes: When a vehicle travels to a preset U-turn section, sampling points are obtained at preset discrete intervals at the center line of the U-turn space or at the boundary line between the current lane and the target lane. The U-turn space is the space between the vehicle's current lane and the opposite lane where a U-turn can be made, and the sampling points are the positions that the vehicle passes through when making a U-turn at the center line or boundary line of the U-turn space. The vehicle adjustment space is constructed based on the sum of the widths of the lane the vehicle is currently in and the oncoming lane, as well as the length of the U-turn space. Based on the vehicle's current location, the sampling point, the vehicle's adjustment space, and the road conditions of the lane to be entered, a vehicle U-turn trajectory is generated using a preset path planning method. If the vehicle's U-turn trajectory cannot be generated based on the sampling points, the driving range that can enter the lane to be entered is expanded, and the sampling point density is reduced to obtain the latest sampling points. The latest sampling point and the driving range are used to perform trajectory planning through the preset path planning method until the vehicle's U-turn trajectory is generated.

2. The method according to claim 1, characterized in that, The step of generating a vehicle U-turn trajectory based on the vehicle's current position, the sampling points, the vehicle's adjustment space, and the road conditions of the lane to be entered using a preset path planning method includes: If there is an obstacle in the lane to be entered, the volume of the obstacle is obtained by predicting its volume based on the center coordinates of the obstacle and its projection. Based on the vehicle's current position, the volume of the obstacle, the sampling point, and the vehicle's adjustment space, a U-turn trajectory is generated by the preset path planning method to bypass the obstacle.

3. The method according to claim 1, characterized in that, The step of generating a vehicle U-turn trajectory based on the vehicle's current position, the sampling points, the vehicle's adjustment space, and the road conditions of the lane to be entered using a preset path planning method includes: The number of vehicles and their speed in the lane to be entered are used to determine whether the lane to be entered is a congested lane. If it is a congested lane, then a lane that is in the same direction as the lane to be entered and is not congested will be identified as the new lane to be entered. The lane entry position is determined in the new lane to be entered based on the vehicle's current position and preset attitude angle. The vehicle's U-turn trajectory is generated using the preset path planning method based on the vehicle's current position, lane entry position, sampling point, and vehicle adjustment space.

4. The method according to claim 3, characterized in that, After the step of determining whether the lane to be entered is a congested lane based on the number of vehicles and their speed in the lane to be entered, the following steps are included: If the lane to be entered is a congested lane, then the lane to be entered is subjected to vertical projection based on the length of the U-turn space to obtain a search point; Based on the search point, search for other vehicles ahead of the lane to be entered and record the search distance. If the search distance is greater than the preset search distance, determine the queuing point based on the last vehicle in the lane to be entered. The vehicle's U-turn trajectory is generated using the preset path planning method based on the vehicle's current location, the queuing point, the sampling point, and the vehicle's adjustment space.

5. The method according to any one of claims 1-3, characterized in that, The step of generating the vehicle U-turn trajectory also includes: If the current lane is a congested lane, the vehicle waiting to make a U-turn is identified by comparing the position and orientation of the vehicle in front. The predicted turning trajectory of the vehicle to be turned around is predicted by a preset trajectory prediction method, and the vehicle turns around according to the predicted turning trajectory.

6. The method according to any one of claims 1-3, characterized in that, After the step of generating the vehicle U-turn trajectory, the method further includes: If other vehicles are detected making a U-turn within the designated U-turn space, a preset neural network is used to predict the U-turn trajectories of these other vehicles to obtain their predicted U-turn paths. Delete the trajectories of several vehicle U-turn trajectories that intersect with the predicted U-turn path.

7. A U-turn trajectory generation device based on autonomous driving, characterized in that, The device includes: The sampling unit is used to acquire sampling points at preset discrete intervals at the center line of the U-turn space or the boundary line between the current lane and the target lane when the vehicle travels to the preset U-turn section. The U-turn space is the space between the lane where the vehicle is currently located and the opposite lane where a U-turn can be made. The sampling points are the positions that the vehicle passes through when making a U-turn on the center line or the boundary line of the U-turn space. A construction unit is used to construct a vehicle adjustment space based on the sum of the widths of the lane where the vehicle is currently located and the oncoming lane, as well as the length of the U-turn space. The planning unit is used to generate a vehicle U-turn trajectory based on the vehicle's current position, the sampling points, the vehicle's adjustment space, and the road conditions of the lane to be entered, using a preset path planning method. If the vehicle U-turn trajectory cannot be generated based on the sampling points, the driving range that can enter the lane to be entered is expanded, and the sampling point density is reduced to obtain the latest sampling points. The latest sampling points and the driving range are used to perform trajectory planning using the preset path planning method until the vehicle U-turn trajectory is generated.

8. A computer device, characterized in that, The computer device includes a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the method as described in any one of claims 1-6.

9. A storage medium, characterized in that, The storage medium stores a computer program, which includes program instructions that, when executed by a processor, can implement the method as described in any one of claims 1-6.

Images