An intersection passing plan method and system, and a storage medium

Abstract

The application provides a method and system for intersection traffic planning and a storage medium. The method comprises the following steps: perceiving intersection information, target obstacle information and self-vehicle running track within a preset range of the self-vehicle, obtaining a predicted track of the target obstacle based on the target obstacle information, obtaining a relevant predicted track based on the predicted track and the intersection information, and controlling the self-vehicle to continue driving or yield based on the relevant predicted track and the self-vehicle running track by using a preset traffic rule. The method provided by the application effectively solves the technical defects that the prior art method for driving an autonomous vehicle at an intersection without traffic signal control depends on a large amount of training data, a complex parameter adjustment process, consumes significant computing resources, has poor interpretability and lacks a formal safety guarantee mechanism.

Description

Technical Field

[0001] This application belongs to the field of autonomous driving technology, specifically relating to an intersection traffic planning method, system, and storage medium. Background Technology

[0002] In the field of autonomous driving technology, intersections lacking traffic signal control remain a significant challenge for autonomous vehicles due to their complex dynamic environment. This is primarily due to the lack of clear right-of-way instructions and the complex interactions with various road users. A key difficulty lies in accurately determining the location and timing of stops to maintain traffic continuity and minimize interference with other road users. Recent research emphasizes that most current methods rely on advanced frameworks such as deep reinforcement learning (DRL), game theory, and probabilistic programming. These methods aim to handle partial observability, infer the intentions of other drivers, and perform real-time risk assessment through predictive modeling and learning-based decision-making strategies.

[0003] However, despite their theoretically strong adaptability and strategy optimization capabilities, these methods face numerous challenges in practical deployment. They typically rely on large amounts of training data, complex parameter tuning processes, and consume significant computational resources. Furthermore, poor interpretability and the lack of formal security mechanisms cast doubt on their reliability in safety-critical scenarios such as unsignalized intersections. Summary of the Invention

[0004] To address the aforementioned technical problems, this application proposes a method, system, and storage medium for intersection traffic planning that features high reliability, low computational resource consumption, and does not rely on large amounts of training data or complex parameter adjustment processes.

[0005] Specifically, this application proposes an intersection traffic planning method, including: The system senses intersection information, target obstacle information, and the vehicle's trajectory within a preset range. Based on the target obstacle information, it obtains the predicted trajectory of the target obstacle. Based on the predicted trajectory and the intersection information, it obtains a related predicted trajectory. Based on the related predicted trajectory and the vehicle's trajectory, it uses preset traffic rules to control the vehicle to continue driving or yield.

[0006] In the above technical solution, the autonomous vehicle is controlled to continue driving or yield by using preset traffic rules based on the vehicle's running trajectory and related predicted trajectory. This enables the autonomous vehicle to make reasonable plans based on the trajectories of other traffic participants at intersections without traffic signals, without relying on a large amount of training data and complex parameter adjustment processes. This significantly reduces the system's computational resource consumption. By sensing the intersection information, target obstacle information, and the vehicle's running trajectory within the preset range of the autonomous vehicle, efficient and low-cost vehicle control is achieved. By obtaining related predicted trajectories from the predicted trajectory, the autonomous vehicle is controlled based on the related predicted trajectories and the vehicle's running trajectory. This makes the system highly interpretable and has a formal safety guarantee mechanism, effectively improving the reliability of autonomous vehicles passing through intersections without traffic signals.

[0007] As one implementation method, the sensing of intersection information, target obstacle information, and vehicle trajectory within a preset range of the self-driving vehicle includes: Determine whether there is intersection information within the preset range of the vehicle. If so, collect the intersection information; otherwise, continue sensing until the intersection information exists within the preset range of the vehicle.

[0008] By collecting intersection information only when it is determined that the vehicle has access to an intersection within a preset range, and then using this information for subsequent intersection traffic planning, unnecessary calculations are avoided on straight roads or roads without intersections, thus saving system computing resources. Continuing to sense the presence of intersection information within the preset range when no intersection information is available ensures continuous attempts until the target is successfully captured, effectively enhancing robustness in real-world, uncertain environments and preventing the entire function from failing due to a single failed sensing attempt.

[0009] Furthermore, the sensing of intersection information, target obstacle information, and vehicle trajectory within a preset range also includes: The vehicle senses all obstacle information within a preset range and filters the obstacle information based on preset filtering rules to obtain the target obstacle information.

[0010] At complex intersections, there may be multiple obstacles, such as vehicles, pedestrians, bicycles, and roadblocks. By filtering these obstacles using preset filtering rules, a large number of irrelevant or low-relevance obstacles can be quickly eliminated, effectively improving computational efficiency. These preset filtering rules also effectively ensure the interpretability of the system.

[0011] Furthermore, obtaining the relevant predicted trajectory based on the predicted trajectory and the intersection information includes: Based on the intersection information, the intersection area is obtained, and it is determined whether the predicted trajectory passes through the intersection area. If so, the predicted trajectory is determined to be a relevant predicted trajectory; otherwise, the predicted trajectory is determined to be an irrelevant predicted trajectory.

[0012] Each detected obstacle has a predicted trajectory, but not every predicted trajectory is related to the vehicle's traffic safety. Therefore, by determining whether the predicted trajectory passes through the intersection area, irrelevant predicted trajectories can be effectively eliminated, which can effectively improve the overall efficiency and accuracy of intersection traffic planning and reduce the system's computational burden and computational resource consumption.

[0013] Furthermore, the step of controlling the vehicle to continue driving or yield based on the relevant predicted trajectory and the vehicle's running trajectory using preset traffic rules includes: The minimum distance between the relevant predicted trajectory and the vehicle's running trajectory is obtained. The minimum distance is compared with a preset distance threshold. If the minimum distance is less than the preset distance threshold, the vehicle is controlled to give way using the preset traffic rules. Otherwise, the vehicle continues to drive along the vehicle's running trajectory.

[0014] By using a preset distance threshold, when the minimum distance between the predicted trajectory of the target obstacle and the trajectory of the vehicle is less than the preset distance threshold, it indicates that there may be a collision risk. Therefore, by controlling the vehicle to give way, the driving safety of the vehicle can be effectively ensured.

[0015] Furthermore, the step of controlling the vehicle to yield using the preset traffic rules includes: Based on the intersection information, obtain all lane information, and then filter the lane information using a preset filtering rule to obtain relevant lane information. Based on the relevant lane information, determine whether the relevant lane intersects with the vehicle lane. If so, obtain all intersection points; otherwise, discard the relevant lane information.

[0016] By analyzing lane information, it is possible to predict which lanes are potential sources of threat before the vehicle even enters the intersection. By filtering all lane information through preset filtering rules, relevant lane information is filtered out, which greatly saves computing resources. By obtaining all intersection points between relevant lane information and the vehicle's lane, dangerous areas can be identified.

[0017] Furthermore, obtaining intersection traffic planning information and controlling vehicle yielding also includes: Project all intersection points onto the vehicle's trajectory, select the intersection point with the closest longitudinal distance to the current vehicle position as a candidate parking position, select the final parking position based on all candidate parking positions and a preset safety distance, and control the vehicle to drive to the final parking position to give way to the target obstacle.

[0018] By selecting the intersection point with the closest longitudinal distance to the current vehicle position among all intersection points as the candidate parking location, the system ensures that the vehicle does not stop prematurely or unnecessarily, thereby significantly reducing unnecessary travel time delays and improving the overall communication efficiency of the intersection. The final parking location is selected based on all the candidate parking locations and a preset safety distance, and the vehicle is controlled to travel to the final parking location, making the behavior of the autonomous vehicle smoother and more predictable by other road users.

[0019] Based on the same inventive concept, this application also proposes a system for intersection traffic planning, the system comprising: The information perception module is used to perceive intersection information, target obstacle information, and the vehicle's trajectory within a preset range.

[0020] The trajectory prediction module is used to obtain the predicted trajectory of the target obstacle based on the target obstacle information.

[0021] The trajectory filtering module is used to obtain relevant predicted trajectories based on the predicted trajectory and the intersection information.

[0022] In addition, a driving planning module is used to control the vehicle to continue driving or give way based on the relevant predicted trajectory and the vehicle's running trajectory using preset traffic rules.

[0023] Furthermore, the driving planning module includes: The yield judgment module is used to obtain the minimum distance between the relevant predicted trajectory and the vehicle's running trajectory, and compare the minimum distance with a preset distance threshold. If the minimum distance is less than the preset distance threshold, the preset passage rule is used to control the vehicle to yield; otherwise, the vehicle continues to be controlled to continue traveling along the vehicle's running trajectory.

[0024] The relevant lane acquisition module is used to acquire all lane information based on the intersection information when controlling the vehicle to yield, and to filter the all lane information using preset filtering rules to obtain relevant lane information.

[0025] The lane intersection determination module is used to determine whether a relevant lane intersects with the vehicle's lane based on the relevant lane information. If so, all intersection points are obtained; otherwise, the relevant lane information is discarded.

[0026] The parking location planning module is used to project all the intersection points onto the vehicle's running trajectory, select the intersection point with the closest longitudinal distance to the current vehicle position as a candidate parking location, and select the final parking location based on all the candidate parking locations and a preset safety distance.

[0027] The vehicle control module is used to control the vehicle to drive to the final parking position in order to give way to the target obstacle.

[0028] Based on the same inventive concept, this application also proposes a computer-readable storage medium storing computer-executable instructions that can be read and executed by a control processor to implement the intersection traffic planning method.

[0029] Compared with the prior art, this application has at least the following beneficial effects: This application uses preset traffic rules to control the autonomous vehicle's continued driving or yielding by using the vehicle's running trajectory and related predicted trajectories. This enables autonomous vehicles to make reasonable plans based on the trajectories of other traffic participants at intersections without traffic signals, without relying on a large amount of training data and complex parameter adjustment processes. This significantly reduces the system's computational resource consumption. By sensing intersection information, target obstacle information, and the vehicle's running trajectory within the preset range of the autonomous vehicle, efficient and low-cost vehicle control is achieved. By obtaining related predicted trajectories from the predicted trajectories, and then controlling the autonomous vehicle based on the related predicted trajectories and the vehicle's running trajectory, the system has high interpretability and a formal safety guarantee mechanism, effectively improving the reliability of autonomous vehicles passing through intersections without traffic signals. Attached Figure Description

[0030] Figure 1 This is a flowchart illustrating the intersection traffic planning method in an embodiment of this application.

[0031] Figure 2 This is a schematic diagram illustrating the traffic planning of a T-junction as shown in the embodiments of this application.

[0032] Figure 3 This is a schematic diagram of a crossroads traffic planning shown in an embodiment of this application.

[0033] Figure 4 This is a schematic diagram illustrating the traffic planning at the intersection merging into the main road, as shown in the embodiments of this application.

[0034] Figure 5 This is a schematic diagram of an intersection traffic planning system shown in an embodiment of this application. Detailed Implementation

[0035] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0036] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or server that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices. Example 1:

[0037] Please refer to Figure 1 The traffic planning method for this intersection mainly includes steps S100 to S400.

[0038] Step S100 includes: sensing intersection information, target obstacle information, and the vehicle's trajectory within a preset range. The preset range can be approximately 20 meters, but is not limited to this. The intersection information can include intersections without traffic lights, T-junctions, etc., and is not restricted here. The target obstacle information can refer to dynamic obstacle information, such as moving vehicles, pedestrians, and animals. The vehicle's trajectory can be a predicted trajectory extending 50 to 60 meters forward from the current position, but is not limited to this.

[0039] Step S200 includes: obtaining the predicted trajectory of the target obstacle based on the target obstacle information. The predicted trajectory of the target obstacle can primarily be its trajectory over the next 5 to 6 seconds. This predicted trajectory can be obtained through kinematic laws, high-precision map prediction, deep learning models, game theory, and other methods, without limitation.

[0040] Step S300 includes: obtaining a relevant predicted trajectory based on the predicted trajectory and the intersection information. Specifically, this can be achieved by obtaining the polygonal region of the intersection based on the intersection information. When the predicted trajectory intersects with the polygonal region, or when the target obstacle is within the polygonal region of the intersection, the predicted trajectory is determined to be a relevant predicted trajectory, and irrelevant predicted trajectories are discarded from all predicted trajectories.

[0041] Step S400 includes: controlling the vehicle to continue driving or yield based on the relevant predicted trajectory and the vehicle's running trajectory using preset traffic rules. Specifically, this involves obtaining the minimum Euclidean distance between the relevant predicted trajectory and the vehicle's running trajectory, comparing this minimum Euclidean distance with a preset distance threshold. If the minimum Euclidean distance is less than the preset distance threshold, it is determined that there is a risk of collision between the vehicle and the target obstacle, and the vehicle is controlled to yield. If the minimum Euclidean distance is greater than or equal to the preset distance threshold, it is determined that there is no risk of collision between the vehicle and the target obstacle, and the vehicle continues to drive. Those skilled in the art can set different preset distance thresholds according to actual needs, without limitation; for example, it can be set to 3 meters, 4 meters, 5 meters, etc.

[0042] In some embodiments, the sensing of intersection information, target obstacle information, and vehicle trajectory within a preset range of the vehicle includes: Determine whether there is intersection information within the preset range of the vehicle. If so, collect the intersection information; otherwise, continue sensing until the intersection information exists within the preset range of the vehicle.

[0043] Those skilled in the art can set different preset ranges for different situations according to actual needs, without any restrictions. For example, when the vehicle speed is high, such as greater than 60 km / h, the preset range can be set to 50 meters, and when the vehicle speed is low, such as 30 km / h, the preset range can be set to 20 meters.

[0044] Preferably, the sensing of intersection information, target obstacle information, and vehicle trajectory within a preset range of the vehicle further includes: The vehicle senses all obstacle information within a preset range and filters the obstacle information based on preset filtering rules to obtain the target obstacle information.

[0045] The obstacle information mainly includes moving vehicles, pedestrians, and animals, as well as stationary roadblocks, trash cans, streetlights, and virtual obstacles caused by sensor noise, false detections, and reflected false targets. Static and virtual obstacles are filtered out using preset filtering rules, retaining only dynamic obstacle information. These preset filtering rules can either eliminate static obstacles based on fixed objects already marked on a high-precision map or track all obstacles in real time. If an obstacle's movement speed is consistently below a fixed value (e.g., less than 0.1 m / s or 0.5 m / s) across multiple frames of data, it is considered a stationary obstacle and eliminated. Information can be fused from multiple sensors, requiring at least one obstacle to be detected by multiple different types of sensors. If not, the obstacle is considered a virtual obstacle and is eliminated.

[0046] Preferably, obtaining the relevant predicted trajectory based on the predicted trajectory and the intersection information includes: Based on the intersection information, the intersection area is obtained, and it is determined whether the predicted trajectory passes through the intersection area. If so, the predicted trajectory is determined to be a relevant predicted trajectory; otherwise, the predicted trajectory is determined to be an irrelevant predicted trajectory.

[0047] The intersection area can primarily be represented as a polygonal region, such as the rectangular area of ​​a crossroads. When the predicted trajectory passes through this rectangular area, it indicates that the target obstacle is about to enter the intersection area or is already in the intersection area. In this case, the predicted trajectory of the target obstacle is determined to be a relevant predicted trajectory, such as the predicted trajectory of vehicles and pedestrians about to cross the intersection, or the predicted path of vehicles or pedestrians currently at the intersection. When the predicted trajectory of the target obstacle does not pass through the intersection area, it is determined to be an irrelevant predicted trajectory, such as the predicted trajectory of a pedestrian walking normally on the crosswalk.

[0048] Preferably, the step of controlling the vehicle to continue driving or yield based on the relevant predicted trajectory and the vehicle's running trajectory using preset traffic rules includes: The minimum distance between the relevant predicted trajectory and the vehicle's running trajectory is obtained. The minimum distance is compared with a preset distance threshold. If the minimum distance is less than the preset distance threshold, the preset traffic rules are used to control the vehicle to give way; otherwise, the vehicle continues to drive along the vehicle's running trajectory.

[0049] The minimum distance can be primarily represented as the length of the line connecting the minimum distance point between the predicted trajectory of the target obstacle and the trajectory of the vehicle. This preset distance threshold can be set according to actual needs and is not limited here. For example, it can be set to 3 meters, 2.5 meters, etc. That is, when the minimum distance between the two curves is lower than this preset distance threshold, it indicates a risk of collision between the two vehicles, and the vehicle should yield. Otherwise, the vehicle should continue to move. The preset passage rules mainly involve determining the relevant lanes based on lane information, obtaining the intersection point between the relevant lanes and the vehicle's lane, and determining the final stopping position based on the intersection point to control the vehicle to stop at the final stopping position to yield to the target obstacle.

[0050] Preferably, controlling the yielding of vehicles using the preset traffic rules includes: Based on the intersection information, obtain all lane information. Based on all lane information, perform filtering processing using preset filtering rules to obtain relevant lane information. Based on the relevant lane information, determine whether the relevant lane intersects with the vehicle lane. If so, obtain all intersection points; otherwise, discard the relevant lane information.

[0051] The lane information mainly includes the current lane, lanes that share the same parent lane or are the leading lane of the vehicle's lane, lanes that are the leading lane of the vehicle's lane, or lanes that are the leading lane of the vehicle's lane; in other words, all lanes at the intersection the vehicle is about to enter. The preset filtering rule mainly involves obtaining the center lines of all lanes based on this information, determining whether the center lines of other lanes intersect with the center line of the vehicle's lane. If they intersect, the lane is considered relevant, and irrelevant lane information is removed.

[0052] Preferably, obtaining intersection traffic planning information and controlling vehicle yielding further includes: Projecting all intersection points onto the vehicle's trajectory, the intersection point with the closest longitudinal distance to the current vehicle position is selected as a candidate parking position. Based on all candidate parking positions and a preset safety distance, a final parking position is selected, and the vehicle is controlled to drive to the final parking position to yield to the target obstacle.

[0053] The preset safety distance can be set according to actual needs and is not limited here; for example, it can be set to 5 meters. This preset safety distance allows the vehicle to buffer against obstacles, effectively avoiding collisions.

[0054] In the specific implementation process, please refer to Figure 2 This scenario involves a car making a left turn at a typical T-junction without traffic signals. The system detects two vehicles simultaneously entering the intersection, with the specific rules being: once the vehicle is determined to be within 20 meters of the intersection, a target obstacle is detected. Figure 2 For two moving vehicles, the predicted trajectory of the target obstacle is obtained. It is determined whether the minimum distance between the predicted trajectory and the vehicle's trajectory is less than a preset distance threshold. If so, the vehicle needs to stop and yield; otherwise, the vehicle does not need to stop and yield. When the vehicle needs to yield, the intersection point with the closest longitudinal distance between the relevant lane centerline and the vehicle's trajectory is obtained as a candidate parking position. The final parking position is selected according to a preset safety distance, and the vehicle stops and yields based on the final parking position and the preset safety distance.

[0055] Please refer to Figure 3This demonstrates another common scenario for autonomous vehicles on the road: navigating intersections. When it's determined that an intersection exists within the vehicle's preset range, a target obstacle exists within the intersection, and the minimum distance between the predicted trajectory of the target obstacle and the vehicle's trajectory is within a preset distance threshold, the vehicle is deemed to need to yield. Since the vehicle needs to stop, a suitable stopping point must be determined. The intersection points of the intersection lanes and the center lines of the vehicle's lanes are obtained. All intersection points are projected onto the vehicle's path, and the intersection closest to the vehicle's current position is selected. A certain safety buffer distance, i.e., a preset safety gap, is added before this point, and this position is used as the final stopping point. The vehicle is controlled to decelerate and stop at this final stopping point to yield to the target obstacle.

[0056] Please refer to Figure 4 This describes a common scenario for autonomous vehicles on the road: a vehicle merging from a secondary ramp onto a main road. When the system detects the vehicle approaching an intersection and a target obstacle is present, it acquires the predicted trajectory of the obstacle. If the minimum distance between the predicted trajectory of the obstacle and the vehicle's trajectory is less than a preset distance threshold, the vehicle must stop to yield. Since the vehicle must stop, a suitable stopping point needs to be determined. The system also acquires the intersection point where the vehicle's lane intersects with the lane at the intersection. Figure 4 Find the intersection of lane A and the lane line of this vehicle, obtain the intersection of the center line of that lane and the center line of the vehicle's lane, project this intersection onto the vehicle's running path, select the intersection closest to the current position of the vehicle, add a certain safety buffer distance before this point, and use this position as the final parking point. Example 2:

[0057] Please refer to Figure 5 This application also proposes a system using the intersection traffic planning method described in Embodiment 1, which mainly includes: an information perception module, a trajectory prediction module, a trajectory filtering module, and a driving planning module.

[0058] The information perception module is used to perceive intersection information, target obstacle information, and the vehicle's trajectory within a preset range. Specifically, it can obtain intersection information, target obstacle information, and the vehicle's trajectory within the preset range using a high-precision map configured on the vehicle.

[0059] The trajectory prediction module is used to obtain the predicted trajectory of the target obstacle based on the target obstacle information. This target obstacle information can primarily include moving vehicles, pedestrians, and animals. The predicted trajectory is mainly the movement trajectory of the target obstacle over the next 5 to 6 seconds.

[0060] The trajectory filtering module is used to obtain relevant predicted trajectories based on the predicted trajectory and the intersection information. These relevant predicted trajectories are mainly those that pass through the intersection or whose target obstacle is already at the intersection.

[0061] Additionally, a driving planning module is used to control the vehicle to continue driving or yield based on the relevant predicted trajectory and the vehicle's driving trajectory using preset traffic rules. These preset traffic rules mainly refer to determining the vehicle's final stopping position based on the intersection of the lane centerline and the vehicle's lane centerline, and controlling the vehicle to stop and yield based on the final stopping position.

[0062] Furthermore, the driving planning module includes: a yield judgment module, a relevant lane acquisition module, a lane intersection judgment module, a parking position planning module, and a vehicle control module.

[0063] The yield judgment module is used to obtain the minimum distance between the relevant predicted trajectory and the vehicle's running trajectory, and compare the minimum distance with a preset distance threshold. If the minimum distance is less than the preset distance threshold, the preset passage rule is used to control the vehicle to yield; otherwise, the vehicle continues to be controlled to continue traveling along the vehicle's running trajectory.

[0064] The relevant lane acquisition module is used to acquire all lane information based on the intersection information when controlling the vehicle to yield, and to filter the all lane information using preset filtering rules to obtain relevant lane information.

[0065] The lane intersection determination module is used to determine whether a relevant lane intersects with the vehicle's lane based on the relevant lane information. If so, all intersection points are obtained; otherwise, the relevant lane information is discarded.

[0066] The parking location planning module is used to project all the intersection points onto the vehicle's running trajectory, select the intersection point with the closest longitudinal distance to the current vehicle position as a candidate parking location, and select the final parking location based on all the candidate parking locations and a preset safety distance.

[0067] The vehicle control module is used to control the vehicle to drive to the final parking position in order to give way to the target obstacle. Example 3:

[0068] This application also proposes a computer-readable storage medium, the computer-readable storage medium comprising: The computer-readable storage medium stores computer-executable instructions, which, when executed by a control processor, implement the intersection traffic planning method described in Embodiment 1.

[0069] The computer-readable storage medium can be implemented entirely or partially by software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented entirely or partially as a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in the computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., a solid-state drive (SSD)).

[0070] In summary, this application effectively addresses the technical shortcomings of existing methods for controlling autonomous vehicles at intersections without traffic signals. These methods rely on large amounts of training data, involve complex parameter adjustments, consume significant computational resources, and suffer from poor interpretability and a lack of formal safety mechanisms. By using the vehicle's trajectory and related predicted trajectories to control the vehicle's continued driving or yielding according to preset traffic rules, the autonomous vehicle can rationally plan its route at intersections without traffic signals based on the trajectories of other traffic participants. This eliminates the need for large amounts of training data and complex parameter adjustments, significantly reducing the system's computational resource consumption. By sensing intersection information, target obstacle information, and the vehicle's trajectory within a preset range, the application achieves efficient and low-cost vehicle control. Furthermore, by obtaining related predicted trajectories from the predicted trajectories and controlling the vehicle based on these trajectories and the vehicle's actual trajectory, the system exhibits high interpretability and possesses a formal safety mechanism, effectively improving the reliability of autonomous vehicles navigating intersections without traffic signals.

[0071] In the several embodiments provided in this application, it will be understood that each block in the flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than those shown in the figures. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved.

[0072] If the aforementioned functions are implemented as software functional modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion 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 an electronic device to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0073] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of this application. It should be understood that the above descriptions are merely specific embodiments of this application and are not intended to limit the scope of protection of this application. In particular, it should be noted that any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application for those skilled in the art.

Claims

1. A method for traffic planning at intersections, characterized in that, include: It can sense intersection information, target obstacle information, and the vehicle's trajectory within a preset range. The predicted trajectory of the target obstacle is obtained based on the target obstacle information; Based on the predicted trajectory and the intersection information, relevant predicted trajectories are obtained; Furthermore, based on the relevant predicted trajectory and the vehicle's running trajectory, preset traffic rules are used to control the vehicle to continue driving or give way.

2. The intersection traffic planning method according to claim 1, characterized in that, The information on intersections, target obstacles, and the vehicle's trajectory within a preset range of the sensing vehicle includes: Determine whether there is intersection information within the preset range of the vehicle. If so, collect the intersection information; otherwise, continue sensing until the intersection information exists within the preset range of the vehicle.

3. The intersection traffic planning method according to claim 2, characterized in that, The sensing of intersection information, target obstacle information, and vehicle trajectory within a preset range also includes: It can sense information about all obstacles within a preset range of the vehicle. The obstacle information is filtered based on preset filtering rules to obtain the target obstacle information.

4. The intersection traffic planning method according to claim 3, characterized in that, The step of obtaining the relevant predicted trajectory based on the predicted trajectory and the intersection information includes: If the predicted trajectory passes through the intersection information, it is determined that the predicted trajectory is a relevant predicted trajectory; otherwise, it is determined that the predicted trajectory is an irrelevant predicted trajectory.

5. The intersection traffic planning method according to claim 4, characterized in that, The step of controlling the vehicle to continue driving or yield based on the relevant predicted trajectory and the vehicle's running trajectory using preset traffic rules includes: Obtain the minimum distance between the relevant predicted trajectory and the vehicle's running trajectory; The minimum distance is compared with a preset distance threshold, and when the minimum distance is less than the preset distance threshold, the preset traffic rules are used to control the vehicle to give way. Otherwise, continue to control the vehicle to continue traveling along the stated vehicle trajectory.

6. The intersection traffic planning method according to claim 5, characterized in that, The method of controlling vehicle yielding using the preset traffic rules includes: All lane information is obtained based on the intersection information; Based on all the lane information, a preset filtering rule is used to filter and obtain relevant lane information; Based on the relevant lane information, determine whether the relevant lane intersects with the vehicle lane; if so, obtain all intersection points. Otherwise, the relevant lane information is removed.

7. The intersection traffic planning method according to claim 6, characterized in that, Obtaining intersection traffic planning information and controlling vehicle yielding also includes: Project all the intersection points onto the vehicle's running trajectory, and select the intersection point with the closest longitudinal distance to the current vehicle position as the candidate parking position; The final parking location is selected based on all the candidate parking locations and the preset safety distance; Control the vehicle to drive to the final parking position to give way to the target obstacle.

8. A system based on the intersection traffic planning method according to any one of claims 1 to 7, characterized in that, The system includes: The information perception module is used to perceive intersection information, target obstacle information, and the vehicle's trajectory within a preset range. The trajectory prediction module is used to obtain the predicted trajectory of the target obstacle based on the target obstacle information; The trajectory filtering module is used to obtain relevant predicted trajectories based on the predicted trajectory and the intersection information; In addition, a driving planning module is used to control the vehicle to continue driving or give way based on the relevant predicted trajectory and the vehicle's running trajectory using preset traffic rules.

9. The intersection traffic planning system according to claim 7, characterized in that, The driving planning module includes: The yielding judgment module is used to obtain the minimum distance between the relevant predicted trajectory and the vehicle's running trajectory, and compare the minimum distance with a preset distance threshold. If the minimum distance is less than the preset distance threshold, the preset passage rule is used to control the vehicle to yield; otherwise, the vehicle continues to drive along the vehicle's running trajectory. The relevant lane acquisition module is used to acquire all lane information based on the intersection information when controlling the vehicle to yield, and to perform filtering processing on all lane information using preset filtering rules to obtain relevant lane information. The lane intersection determination module is used to determine whether a relevant lane intersects with the vehicle's lane based on the relevant lane information. If so, all intersection points are obtained; otherwise, the relevant lane information is discarded. The parking location planning module is used to project all the intersection points onto the vehicle's running trajectory, select the intersection point with the closest longitudinal distance to the current vehicle position as a candidate parking location, and select the final parking location based on all the candidate parking locations and a preset safety distance. The vehicle control module is used to control the vehicle to drive to the final parking position in order to give way to the target obstacle.

10. A computer-readable storage medium storing computer-executable instructions, characterized in that, When the computer-executable instructions are executed by the control processor, they implement the intersection traffic planning method as described in any one of claims 1 to 7.

Images