Efficient Planning Method and System for Vehicle Queue Trajectories at Intelligent Connected Mixed Traffic Intersections

Abstract

The present invention proposes an efficient method and system for planning vehicle queue trajectories at an intelligent connected and mixed traffic intersection, belonging to the field of intelligent transportation technology. The method includes: constructing a control area based on signal timing information and vehicle information; constructing an initial segmented trajectory by passing through the intersection at the maximum speed limit, and adjusting the trajectory according to the safety boundary constraint to obtain candidate segmented trajectories; if the candidate segmented trajectory arrives at the intersection during the red light period, then based on the start time of the next green light: for CAVs, reversely construct a uniformly accelerating section and connect it with the candidate segmented trajectory through a connecting section, and for HVs, reversely construct a parking section and connect it with the candidate segmented trajectory through a connecting section to obtain the final segmented trajectory; the connecting section includes a uniformly decelerating section, or a combination of a uniformly decelerating section and a parking section. By considering the influence of HV trajectories on CAV trajectory planning, the entire mixed traffic flow passes through the intersection with a smooth trajectory, improving the traffic capacity of the intersection.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent transportation, and particularly to an efficient method and system for planning vehicle queue trajectories at an intelligent connected mixed traffic intersection. Background Art

[0002] With the development of vehicle-road cooperation and intelligent connected vehicle (CAV) technology, the programmable characteristics of CAVs provide new ideas for realizing the collaborative optimization of vehicle trajectories and traffic signals. However, in the intersection scenario where CAVs and human-driven vehicles (HVs) are mixed, existing trajectory planning methods still face significant challenges. On the one hand, the penetration rate of CAVs needs a long-term transition (expected to reach 100% by 2060), and the trajectory coordination of mixed traffic flows needs to be compatible with the uncertainty of HV behavior; on the other hand, the coupling of intersection signal control and vehicle dynamics constraints makes trajectory planning need to meet the requirements of safety, traffic efficiency and real-time simultaneously.

[0003] The existing technologies have the following defects in vehicle trajectory planning at mixed traffic intersections:

[0004] (1) Insufficient computational complexity and real-time performance: For example, the invention patent with the application number 2021103149821 indirectly optimizes vehicle acceleration by calculating the passing time at conflict points, and needs to frequently solve multi-variable non-linear equations, resulting in a sharp increase in the computational burden in the multi-vehicle mixed traffic scenario; while the invention patent with the application number 2021104898377 relies on a global optimization model of signal timing and lane sharing, which requires high-precision numerical iteration and is difficult to meet the on-line control requirements.

[0005] (2) Difficulty in simultaneously handling the trajectory planning problems of multiple vehicles: Since existing methods usually divide vehicles into fleets and only plan the trajectories of the leading vehicles in the fleets, defaulting that the following vehicles follow the leading vehicles in the fleets, they fail to simultaneously handle the trajectory planning problems of multiple vehicles.

[0006] (3) Lack of systematicness in trajectory coordination: For segmented trajectories in existing solutions, no unified analytical solution framework is established, making it difficult to ensure the overall smoothness of the vehicle queue and the matching accuracy with the green light window, and easily causing secondary conflicts or traffic efficiency losses.

[0007] In addition, although the traditional optimal control model can theoretically describe the trajectory planning problem, its defect of relying on numerical solutions and being unable to generate analytical solutions leads to low algorithm efficiency and can only be applied to single-vehicle or small-scale fleet control, and cannot support traffic flow-level collaborative optimization. In the real background of the long-term coexistence of CAVs and HVs, there is an urgent need for a trajectory planning method with low computational complexity and high real-time performance to balance the dynamic adaptability of HV behavior and the collaborative efficiency of CAV queues, so as to break through the bottleneck of existing technologies and achieve a systematic improvement in intersection traffic capacity. Summary of the Invention

[0008] To solve the problem that the existing vehicle trajectory control methods at intersections only focus on the vehicle dimension and it is difficult to plan the trajectories of all CAVs simultaneously from the overall dimension of traffic flow to make them pass through the intersection smoothly, the present invention proposes an efficient method and system for planning the vehicle queue trajectories at intelligent connected mixed traffic intersections. By decomposing the CAV trajectory into several smooth curves and constructing the safety boundary of the CAV trajectory based on the Newell following model; on this basis, an analytical solution method for each curve segment is given to achieve the rapid planning of the CAV trajectory. For HVs in the mixed traffic flow, a rapid estimation method for the HV trajectory curve is proposed, and the influence of the HV trajectory on the CAV trajectory planning is considered, so that the entire mixed traffic flow passes through the intersection with a smooth trajectory, thereby efficiently using the green light window to ensure the safe and efficient passage of the vehicle queue through the intersection and improving the intersection capacity.

[0009] To achieve the above object, the present invention adopts the following technical solutions:

[0010] In the first aspect, the present invention provides an efficient method for planning the vehicle queue trajectories at intelligent connected mixed traffic intersections, including:

[0011] Construct a control area based on the signal timing information, the initial speeds and initial times of the intelligent connected vehicles and the human-driven vehicles entering the control area.

[0012] In the control area, construct an initial segmented trajectory based on passing through the intersection at the maximum speed limit, and adjust the trajectory according to the safety boundary constraint to obtain a candidate segmented trajectory.

[0013] If the candidate segmented trajectory arrives at the intersection during the red light period, then based on the next green light start time: for the intelligent connected vehicle, reversely construct a uniformly accelerated segment and connect it with the candidate segmented trajectory through a connecting segment to obtain the final segmented trajectory; for the human-driven vehicle, reversely construct a parking segment and connect it with the candidate segmented trajectory through a connecting segment to obtain the final segmented trajectory; the connecting segment includes a uniformly decelerated segment, or a combination of a uniformly decelerated segment and a parking segment.

[0014] Store the final segmented trajectory of the current vehicle in the queue, and when subsequent vehicles enter the control area, iteratively update the trajectory information to achieve trajectory planning.

[0015] In the second aspect, the present invention provides an efficient system for planning the vehicle queue trajectories at intelligent connected mixed traffic intersections, including:

[0016] A scenario construction module configured to construct a control area based on the signal timing information, the initial speeds and initial times of the intelligent connected vehicles and the human-driven vehicles entering the control area.

[0017] The candidate trajectory construction module is configured to construct an initial segmented trajectory based on passing through the intersection at the highest speed limit, and adjust the trajectory according to the safety boundary constraint to obtain a candidate segmented trajectory;

[0018] The final trajectory construction module is configured to, if the candidate segmented trajectory arrives at the intersection during the red light period, based on the next green light start time: for the connected and automated vehicle, construct a uniformly accelerated section in reverse and connect it to the candidate segmented trajectory through a connection section to obtain a final segmented trajectory; for the human-driven vehicle, construct a parking section in reverse and connect it to the candidate segmented trajectory through a connection section to obtain a final segmented trajectory; the connection section includes a uniformly decelerated section, or a combination of a uniformly decelerated section and a parking section;

[0019] The trajectory update module is configured to store the final segmented trajectory of the current vehicle in a queue, and when a subsequent vehicle enters the control area, iteratively update the trajectory information to achieve trajectory planning.

[0020] In a third aspect, the present invention provides a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, the steps in an efficient method for planning the trajectory of a vehicle queue at an intersection of connected and automated vehicles as described in the first aspect are implemented.

[0021] In a fourth aspect, the present invention provides a computer device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, and when the processor executes the program, the steps in an efficient method for planning the trajectory of a vehicle queue at an intersection of connected and automated vehicles as described in the first aspect are implemented.

[0022] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0023] Based on the Newell car-following model, the present invention constructs the safety boundary of the CAV trajectory, decomposes the CAV trajectory into several curves that can be solved analytically, and establishes a solution method for each trajectory curve. At the same time, the trajectory estimation of the human-driven vehicle (HV) is realized, and together with the signal timing plan, it is used as the input for the CAV trajectory planning. Different from the existing intersection vehicle trajectory control methods, the vehicle trajectory planning method proposed by the present invention has high computational efficiency and the characteristics of being able to handle the trajectory planning problems of a series of vehicles simultaneously, and can be fully applied to the online control of intersection vehicle trajectories.

[0024] The advantages of the additional aspects of the present invention will be partially given in the following description, partially become obvious from the following description, or be understood through the practice of the present invention. Description of the Drawings

[0025] The accompanying drawings forming a part of the present invention are used to provide a further understanding of the present invention. The schematic embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute a limitation to the present invention.

[0026] Figure 1 It is the main flowchart of an efficient vehicle queue trajectory planning method for an intelligent connected mixed traffic intersection provided by an embodiment of the present invention;

[0027] Figure 2 It is the trajectory planning flowchart provided by an embodiment of the present invention;

[0028] Figure 3 It is the Newell following model provided by an embodiment of the present invention;

[0029] Figure 4 It is the schematic diagram of the initial segmented trajectory provided by an embodiment of the present invention;

[0030] Figure 5 It is the schematic diagram of the correction of the initial segmented trajectory provided by an embodiment of the present invention;

[0031] Figure 6 It is the schematic diagram of the CAV final segmented trajectory estimation provided by an embodiment of the present invention; wherein, (a) shows the case where the uniformly accelerated segment constructed in reverse is relatively close to the candidate segmented trajectory; (b) shows the case where the uniformly accelerated segment constructed in reverse is relatively far from the candidate segmented trajectory;

[0032] Figure 7 It is the schematic diagram of the HV trajectory estimation provided by an embodiment of the present invention;

[0033] Figure 8 It is the schematic diagram of the CAV arrival time calculation provided by an embodiment of the present invention;

[0034] Figure 9 It is the schematic diagram of the trajectory node state calculation provided by an embodiment of the present invention. Detailed implementation manners

[0035] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0036] At signalized intersections, due to the alternation of traffic signals between green and red lights, the traffic flow on the signal trunk line is usually forced to suddenly decelerate and accelerate. When the traffic volume is large, a stop-and-go traffic pattern will form. This stop-and-go traffic has many adverse effects on the traffic flow. Obviously, frequent stops and starts will result in a poor experience for drivers and passengers, and significantly increase the risk of traffic conflicts, and also increase the fuel consumption and emissions of vehicles. In addition, when the vehicle decelerates or stops, it will cause a decrease in traffic capacity, further leading to an increase in travel time and delay.

[0037] Although the stop-and-go traffic phenomenon has been widely studied through theoretical models and empirical observations in the context of highway traffic, few studies have explored how to effectively alleviate traffic congestion and its adverse effects on signal-controlled roads before the emergence of vehicle-road cooperation and autonomous driving technologies. It can be said that before the emergence of Connected Automated Vehicles (CAVs), the dynamic behavior of vehicles was mainly determined by the microscopic behavior of human drivers. However, due to the unpredictability of human behavior and the lack of empirical data to comprehensively describe such behavior, there is no widely accepted human driving behavior model. Therefore, traditional infrastructure-based control means (such as traffic signals) are difficult to perfectly smooth vehicle trajectories. These control means are designed to adapt to human behavior, but their effectiveness is limited by the uncertainty of human behavior.

[0038] In contrast, CAVs can (at least partially) replace human drivers with programmable robots, whose driving algorithms can be flexibly customized and precisely executed. This makes it possible to achieve coordination between the trajectories of individual vehicles and the overall infrastructure-based control, thus optimizing the driving experience and overall traffic performance. These potentials have inspired a series of pioneering studies, such as exploring how to use CAVs to improve the traffic efficiency and safety at intersections and reduce the environmental impact of highway sections.

[0039] However, most existing studies have formulated the vehicle trajectory planning problem as an optimal control model. However, optimal control models usually cannot give analytical solutions, and existing control algorithms rely on complex numerical calculations. Even so, the solution efficiency of optimal control models is not high and it is difficult to meet the online control of a large number of control objectives. For this reason, the control objects of existing controls are limited to controlling one or a few vehicles in specific highway facilities (such as intersections or sections) to achieve specific goals (such as stability, safety or fuel efficiency), rather than improving the overall traffic performance by smoothing the vehicle flow. Therefore, how to achieve efficient vehicle trajectory control with low computational complexity remains an important challenge in current research.

[0040] A smooth driving trajectory means that vehicles will be safer when passing through intersections, with less fuel consumption and emissions, and a better riding experience. For each vehicle, its trajectory has control points in an infinite number of dimensions, and there are also complex interactions between vehicles. This makes it an extremely complex task to simultaneously plan the trajectories of all CAVs within the control area. On the other hand, the popularization of CAVs is a long-term task and a continuous development process. Some studies predict that the penetration rate of CAVs may not reach 100% until around 2060. In this environment, accurately estimating the driving trajectory of human-driven vehicles (HV) is an essential task for CAV trajectory planning, which makes this problem even more challenging.

[0041] Therefore, the following embodiments provide an efficient method, system, medium, and device for planning the trajectories of vehicle queues at intelligent connected mixed traffic intersections. For different types of vehicles (HV and CAV), different trajectory planning strategies are adopted, which can give full play to the characteristics of various vehicles, enable vehicles to pass through intersections smoothly, thereby effectively improving the traffic efficiency of intersections, reducing vehicle delays and energy consumption, and ensuring smooth traffic flow.

[0042] Embodiment 1

[0043] As Figure 1 shown, this embodiment discloses an efficient method for planning the trajectories of vehicle queues at intelligent connected mixed traffic intersections, including the following steps:

[0044] S1. Based on the signal timing information, the initial speeds and initial times of intelligent connected vehicles and human-driven vehicles entering the control area, construct the control area;

[0045] S2. In the control area, based on constructing an initial segmented trajectory passing through the intersection at the maximum speed limit, adjust the trajectory according to the safety boundary constraints to obtain a candidate segmented trajectory;

[0046] S3. If the candidate segmented trajectory arrives at the intersection during the red light period, then based on the start time of the next green light: for intelligent connected vehicles, reversely construct a uniformly accelerating section and connect it to the candidate segmented trajectory through a connection section to obtain the final segmented trajectory; for human-driven vehicles, reversely construct a parking section and connect it to the candidate segmented trajectory through a connection section to obtain the final segmented trajectory; the connection section includes a uniformly decelerating section, or a combination of a uniformly decelerating section and a parking section;

[0047] S4. Store the final segmented trajectory of the current vehicle in the queue. When subsequent vehicles enter the control area, iteratively update the trajectory information to achieve trajectory planning.

[0048] Next, in combination with Figure 1 , a detailed description will be given of an efficient method for planning the trajectories of vehicle queues at intelligent connected mixed traffic intersections disclosed in this embodiment.

[0049] (1) Intersection Vehicle Trajectory Planning System Architecture

[0050] Consider a typical signalized intersection with a control area of length L. At each approach, the starting position of the control area is 0 and the position of the downstream stop line is L. The speed limit at the intersection is , and the maximum acceleration and maximum deceleration of the vehicle are and respectively. The start time of each control cycle is the zero moment, and the start time of the green phase is S and the green time is G. Then the green phase can be expressed as .

[0051] Within each signal cycle, the central controller plans the driving trajectory for each CAV according to the initial state of the CAV entering the control area and the signal timing information.

[0052] Specifically: within the control cycle c, when the vehicle enters the control area, the intersection central controller can obtain its initial speed and the time when it enters the control area. The central controller plans the trajectory of the CAV or estimates the trajectory of the HV according to the initial state of the vehicle and the signal timing information, and stores it in the vector P, , where N is the set of vehicles in the platoon queue. At the same time, the central controller will send the planned trajectory curve to the corresponding CAV, enabling it to safely pass through the intersection at a speed during the green light period, effectively improving the utilization efficiency of the green light time.

[0053] It should be noted that the trajectory planning method proposed in the present invention is also applicable to the distributed control mode. Specifically: when the CAV enters the control area, it will receive the trajectory information of the vehicle in front and the traffic signal timing information, and it can calculate the trajectory curve by relying on its own computer and send it to the central controller.

[0054] The trajectory planning methods for left-turning vehicles and straight-going vehicles are the same, and the specific process is as shown in Figure 2 .

[0055] (2) CAV Trajectory Planning Method

[0056] The trajectory of vehicle n is defined as a piecewise quadratic function , which is specifically composed of one or several trajectory segments such as a uniformly accelerating segment with an acceleration of , a uniformly decelerating segment with a deceleration of , a constant-speed segment, and a stopping segment connected together, and the two connected trajectory curves are tangent at the connection point, that is, they have the same speed.

[0057] The goal of the CAV trajectory planning algorithm proposed by the present invention is to construct a smooth vehicle trajectory while considering the safety of the vehicle during the following process. According to the Figure 3 shown Newell following model, the trajectory of the following vehicle n can be obtained by translating the trajectory of the leading vehicle in time and space, as shown in the following formula:

[0058]

[0059] In the formula, is the vehicle perception reaction time, is the minimum spacing after the vehicle makes an emergency brake; Figure 3 in represents the headway.

[0060] Therefore, in order to ensure the safety of the vehicle during the following process, the vehicle must not exceed the safety boundary at any time t. According to the Newell following model, the safety boundary curve of vehicle n is given by the following formula:

[0061]

[0062] Generally, vehicles always tend to pass through the intersection in the fastest way to obtain the maximum traffic efficiency. That is to say, when the vehicle enters the intersection control area at the initial speed , in the case of no signal lights and no influence from the leading vehicle, it will always accelerate with the acceleration to the speed limit value , and maintain the speed to pass through the intersection. Therefore, CAV trajectory planning needs to solve two main problems: 1) whether it is affected by the leading vehicle; 2) the signal light state when arriving at the intersection. The specific process is as follows:

[0063] Step 1) Generate the trajectory for the vehicle to reach the intersection as fast as possible under ideal conditions

[0064] First, construct a uniformly accelerated trajectory segment that maintains the acceleration starting from the initial state point . The end point coordinates of the acceleration segment trajectory are given by formula (3). Then, starting from the point, the vehicle travels at a constant speed until it reaches the intersection stop line. As Figure 4 shown, the ideal fastest arrival trajectory consists of trajectory segment ① and trajectory segment ②, and this is used as the initial segmented trajectory .

[0065]

[0066] Step 2) Adjust the trajectory according to whether the fastest driving trajectory is affected by the vehicle ahead

[0067] According to the trajectory curve of the vehicle ahead , generate a safety boundary curve using formula (2) . Establish the following quadratic equation:[[]]

[0068]

[0069] If the equation has no real solutions, it is a candidate trajectory and does not need to be corrected; otherwise, if there are real solutions, it can be determined that the initial segmented trajectory intersects with the safety boundary . At this time, the initial segmented trajectory needs to be corrected so that it decelerates at the maximum deceleration at an appropriate position and finally smoothly cuts into the safety boundary, as Figure 5 shown. In this case, the candidate trajectory after the starting deceleration position consists of two parts, namely the merging trajectory segment ③ and the safety boundary ④ after the merging point.[[]]

[0070] It should be understood that for determining the "appropriate position" to insert the constant deceleration segment and for determining all nodes in this embodiment, they are all implemented by the method in "(5) Trajectory segment endpoint calculation method" below.[[]]

[0071] Step 3) Further adjust the trajectory according to the state of the candidate trajectory when it reaches the intersection to form the final trajectory

[0072] Use to represent the generalized inverse function of the trajectory function , and its definition is given by formula (5).[[]]

[0073]

[0074] When the candidate trajectory reaches the intersection within the green light phase, that is, when it satisfies , there is no need to adjust the candidate trajectory, and it can be directly used as a feasible trajectory of vehicle n .

[0075] On the contrary, if the candidate trajectory reaches the intersection during the red light time (that is ), then the candidate trajectory needs to be adjusted.[[]]

[0076] As Figure 6 (a) shown, first translate the part of the candidate trajectory with a position greater than L to the right until the next green light start time, and then reverse-construct a trajectory that maintains the maximum acceleration The acceleration segment trajectory ⑥ until it is close enough to the candidate trajectory Finally, from a suitable position on the candidate trajectory Construct a deceleration trajectory segment ⑤ with a deceleration of until it is tangent to the trajectory segment ⑥.

[0077] In addition, as Figure 6 Shown in (b), when the distance between the trajectory segment ⑤ and the candidate trajectory is far, a parking trajectory segment ⑦ needs to be added between the trajectory segment ⑤ and the trajectory segment ⑥. In this way, by smoothly connecting the trajectory after adjusting the arrival time with the candidate trajectory, the final segmented trajectory of vehicle n is obtained.

[0078] (3) HV Trajectory Estimation Method

[0079] In the present invention, the trajectory of the preceding vehicle is an important basis for the CAV trajectory planning. Therefore, the intersection central controller needs to estimate the HV trajectory, and its method is similar to the CAV trajectory planning method.

[0080] Specifically, the HV trajectory estimation also includes three steps: 1) Generate the trajectory of the vehicle arriving at the intersection fastest under ideal conditions; 2) Adjust the trajectory according to whether the fastest driving trajectory is affected by the vehicle in front; 3) Further adjust the trajectory according to the state of the candidate trajectory arriving at the intersection.

[0081] Among them, steps 1) and 2) are exactly the same as steps 1) and 2) in the CAV trajectory planning. The following mainly describes step 3).

[0082] Step 3): Similarly, if the candidate trajectory arrives at the intersection during the green light period, that is, when is satisfied, the candidate trajectory does not need to be corrected and can be directly used as the estimated trajectory of the HV .

[0083] On the contrary, if the candidate trajectory arrives at the intersection during the red light time, the candidate trajectory needs to be adjusted. In this case (that is, ), since the HV needs to stop and wait at the intersection. As Figure 7 shown, first, a trajectory segment with an initial speed of 0 and accelerating with the maximum acceleration needs to be constructed starting from the next green light start time (position L); subsequently, a horizontal trajectory segment ⑦ (parking segment) is also constructed in the reverse direction starting from the next green light start time (position L) until a specific parking point; finally, a merging trajectory segment ⑥ is constructed to smoothly connect the candidate trajectory (① - ⑤, ⑤ is the uniform speed segment) with the trajectory segment ⑦. In this way, the estimated trajectory of the HV is obtained 。

[0084] This embodiment uses a trajectory segment with a fixed acceleration to estimate the trajectory of the HV. However, since the trajectory of the HV is uncontrolled, during the actual car-following process, the HV usually does not accelerate or decelerate at a fixed acceleration or deceleration (such as , ). According to the Newell car-following model, the acceleration of the HV during the car-following process can be given by Equation (6), where . Obviously, its acceleration is not fixed, but is jointly determined by multiple factors. This means that there are certain errors in the proposed HV trajectory estimation method.

[0085]

[0086] According to Equation (6), the acceleration (deceleration) of the HV during the car-following process is always between the maximum deceleration and the maximum acceleration . To reduce this estimation deviation, relatively small accelerations and decelerations are adopted when estimating the HV trajectory, that is, , where .

[0087]

[0088] (4) CAV expected arrival time calculation

[0089] In the CAV and HV mixed traffic environment, the trajectory of the CAV may be affected by the HV in front. As Figure 8 shown, since the HV arriving at the intersection during the red light needs to stop and wait, and starts to accelerate through the intersection from a stationary state after the green light is on; when the rear CAV arrives at the intersection at the maximum speed , the HV may not have accelerated to the maximum speed . At this time, if the distance between the two vehicles is already relatively close, the CAV must decelerate to maintain a safe distance from the HV in front, which will disrupt the stability of the traffic flow behind the CAV, causing certain safety hazards and an increase in vehicle energy consumption.

[0090] To avoid the above situation, it is necessary to postpone the arrival time of the CAV at the intersection by a certain time , so that the trajectory of the CAV can smoothly merge into its safety boundary . This embodiment calls this postponed arrival time the expected arrival time, denoted as .

[0091] Trajectory on the point represents the state of the HV in front passing through the intersection, and the speed of this point is After this point, HV continues to accelerate at the maximum acceleration until it reaches the maximum speed at the point . In this embodiment, the point is translated to the right by and translated downward by to obtain the tangent point of the CAV trajectory and the safety boundary . Therefore,[[]] and and can be given by formula (8) respectively. Accordingly, the determination condition for whether the arrival time needs to be adjusted in step 3) of the CAV trajectory planning should be updated to formula (9).

[0092]

[0093] (5) Method for calculating the endpoints of the trajectory segment

[0094] The vehicle trajectory planning method proposed by the present invention uses fixed control parameters , and the vehicle trajectory curve function can be uniquely determined by the tangent points of each trajectory segment. Therefore, the main work of CAV trajectory planning and HV trajectory estimation lies in calculating the states of the endpoints of each trajectory segment, and the calculation methods for all endpoints are the same. Here, the endpoint calculation in the candidate trajectory correction process of step 2) in the CAV trajectory planning method is taken as an example for illustration.[[]]

[0095] To facilitate the description of the calculation method, some basic concepts and definitions are given below.[[]]

[0096] Definition 1: Any point p on the trajectory can be represented as a tuple consisting of three elements ( ), where and represent the time and position of node p respectively, and represents the speed of node p, that is, the slope of the tangent line at node p.[[]]

[0097] Definition 2: The vehicle trajectory is composed of segments of the trajectory smoothly connected in sequence. Therefore,[[]] can be represented as .

[0098] Adjacent trajectory segments in the vehicle trajectory curve have the same position and speed at the connection point. As shown in Figure 9 , the curve represents a certain segment of the safety boundary of vehicle n, and the curve represents the forward trajectory For a certain section, the starting and ending points of the two trajectories are both known. Assume that The starting time of the merging trajectory segment is , and the two trajectories are tangent at point A ( ). Then the two trajectories have the same position and velocity at the tangent point m. Through arrangement, this embodiment can obtain a binary quadratic equation system about and (the position of node A), which is given by formula (10). By solving the equation system, the values of and can be obtained, and then the position and velocity of the tangent point can be obtained.

[0099]

[0100] Among them, the subscript n represents the current vehicle information, and the subscript n - 1 represents the information of the vehicle in front. For example represents the acceleration of the current vehicle, represents the acceleration of the vehicle in front.

[0101] It should be noted that the equation system (10) may have no solution, or even if there is a solution, it may be invalid. In this case, it means that these two trajectories cannot be tangent, and it is necessary to find a feasible solution by traversing the remaining trajectory segments. The specific calculation process is as follows:

[0102]

[0103] Among them, and represent the starting time and ending time of the trajectory segment , is the ending time of the trajectory segment .

[0104] This specific embodiment simplifies the complex trajectory planning problem to the solution of each node by giving a unified calculation method for the nodes between the segmented curves, greatly reducing the computational complexity, enabling this solution to quickly process a large number of vehicle trajectory planning tasks and meet the requirements of online control. At the same time, the trajectory planning process is meticulous and scientific, optimizing the trajectory step by step in three steps. It not only considers the initial situation where the vehicle passes through the intersection at the maximum speed limit, but also adjusts the trajectory according to the safety boundary constraints, and finally further adjusts according to the signal light state. For HV, by dynamically adjusting its acceleration to adapt to the trajectory estimation error, the accuracy of the planning is improved. In addition, the CAV expected arrival time delay mechanism involved effectively avoids conflicts with the HV in front, ensures the stability of the traffic flow, reduces potential safety hazards and energy consumption. Overall, this embodiment improves the intersection capacity, optimizes the traffic flow, reduces vehicle delays, and provides strong support for the development of intelligent transportation.

[0105] Embodiment 2

[0106] This embodiment provides an efficient planning system for vehicle queue trajectories at an intelligent connected mixed traffic intersection, including:

[0107] A scenario construction module, configured to construct a control area based on signal timing information, the initial speeds and initial times of intelligent connected vehicles and human-driven vehicles entering the control area;

[0108] A candidate trajectory construction module, configured to construct an initial segmented trajectory by passing through the intersection at the maximum speed limit, and adjust the trajectory according to safety boundary constraints to obtain a candidate segmented trajectory;

[0109] A final trajectory construction module, configured to, if the candidate segmented trajectory arrives at the intersection during the red light period, based on the start time of the next green light: for intelligent connected vehicles, construct a uniformly accelerating section in reverse and connect it to the candidate segmented trajectory through a connection section to obtain a final segmented trajectory; for human-driven vehicles, construct a parking section in reverse and connect it to the candidate segmented trajectory through a connection section to obtain a final segmented trajectory; the connection section includes a uniformly decelerating section, or a combination of a uniformly decelerating section and a parking section;

[0110] A trajectory update module, configured to store the final segmented trajectory of the current vehicle in the queue, and when subsequent vehicles enter the control area, iteratively update the trajectory information to achieve trajectory planning.

[0111] Embodiment 3

[0112] This embodiment provides a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, it implements the steps in an efficient planning method for vehicle queue trajectories at an intelligent connected mixed traffic intersection as described in Embodiment 1 above.

[0113] Embodiment 4

[0114] This embodiment provides a computer device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, it implements the steps in an efficient planning method for vehicle queue trajectories at an intelligent connected mixed traffic intersection as described in Embodiment 1 above.

[0115] The steps or modules involved in Embodiments 2 to 4 above correspond to those in Embodiment 1, and the specific implementation manners can be referred to the relevant description part of Embodiment 1. The term "computer-readable storage medium" should be understood to include a single medium or multiple media including one or more instruction sets; it should also be understood to include any medium that can store, encode, or carry an instruction set for execution by a processor and enable the processor to execute any method in the present invention.

[0116] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. An efficient method for planning the vehicle queue trajectory at an intelligent networked mixed traffic intersection, characterized in that, Including: Construct a control area based on signal timing information, the initial speeds and initial times of connected and automated vehicles and human-driven vehicles entering the control area; In the control area, construct an initial segmented trajectory based on passing through the intersection at the highest speed limit, and adjust the trajectory according to safety boundary constraints to obtain a candidate segmented trajectory; If the candidate segmented trajectory arrives at the intersection during the red light period, then based on the start time of the next green light: for connected and automated vehicles, construct a uniformly accelerating section in reverse and connect it to the candidate segmented trajectory through a connecting section to obtain the final segmented trajectory; for human-driven vehicles, construct a parking section in reverse and connect it to the candidate segmented trajectory through a connecting section to obtain the final segmented trajectory; the connecting section includes a uniformly decelerating section, or a combination of a uniformly decelerating section and a parking section; For human-driven vehicles, dynamically adjust the accelerations of their uniformly accelerating section and uniformly decelerating section to adapt to the trajectory estimation error of human-driven vehicles in the mixed traffic flow, specifically expressed as: ； Among them, and represent the maximum acceleration and the maximum deceleration, is a coefficient; In a CAV and HV mixed traffic environment, the trajectory of the CAV may be affected by the HV in front. Since the HV arriving at the intersection during the red light needs to stop and wait, and starts to accelerate through the intersection from a stationary state after the green light turns on; when the CAV behind arrives at the intersection at the maximum speed, the HV may not have accelerated to the maximum speed yet. At this time, if the distance between the two vehicles is already relatively close, the CAV must decelerate to maintain a safe distance from the HV in front, which will disrupt the stability of the traffic flow behind the CAV, and the arrival time of the CAV at the intersection needs to be postponed by a certain amount of time , so that the trajectory of the CAV can smoothly merge into its safety boundary , and the postponed arrival time is called the expected arrival time, denoted as , Among them, is the delay time for the CAV to reach the intersection, is the maximum speed, is the trajectory at the speed at a point, is the acceleration of the HV trajectory; is the vehicle perception and reaction time, is the minimum distance after the vehicle makes an emergency brake, is the desired arrival time, and L is the length of the control area; Store the final segmented trajectory of the current vehicle in a queue, and when subsequent vehicles enter the control area, iteratively update the trajectory information to achieve trajectory planning.

2. The efficient trajectory planning method for vehicle queues at an intelligent networked mixed traffic intersection according to claim 1, wherein, The uniformly accelerating section is for the vehicle to accelerate at the maximum acceleration, and the uniformly decelerating section is for the vehicle to decelerate at the maximum deceleration.

3. The efficient trajectory planning method for vehicle queues at an intelligent networked mixed traffic intersection according to claim 1, wherein The segmented trajectory is a piecewise quadratic function, and adjacent trajectory segments are tangent at the connection point, having the same position and speed; among them, the position and speed of the connection point are solved by establishing a binary quadratic equation system.

4. The efficient trajectory planning method for vehicle queues at an intelligent networked mixed traffic intersection according to claim 1, wherein The constructing an initial segmented trajectory based on passing through the intersection at the highest speed limit and adjusting the trajectory according to safety boundary constraints to obtain a candidate segmented trajectory specifically includes: Construct an initial segmented trajectory based on a uniformly accelerating section and a uniformly moving section at the highest speed limit; Generate a safety boundary curve based on the trajectory of the vehicle ahead. If there is an intersection between the initial segmented trajectory and the safety boundary curve, then insert a uniformly decelerating section into the initial segmented trajectory to obtain a candidate segmented trajectory; otherwise, directly use it as the candidate segmented trajectory.

5. The efficient trajectory planning method for vehicle queues at an intelligent networked mixed traffic intersection according to claim 1, wherein, Based on the generalized inverse function of the candidate segmented trajectory, judge whether the candidate segmented trajectory can pass through the intersection during the current green light phase. If it can, output it as the final segmented trajectory; otherwise, set the trajectory passing time at the start time of the next green light based on the generalized inverse function.

6. The efficient trajectory planning method for vehicle queues at an intelligent networked mixed traffic intersection according to claim 1, characterized in that For connected and automated vehicles, if the uniformly accelerating section constructed in reverse cannot be connected to the candidate segmented trajectory through the uniformly decelerating section, then add a parking section after the uniformly decelerating section to connect to the candidate segmented trajectory; For human-driven vehicles, the connecting section is a uniformly decelerating section.

7. An intelligent networked mixed traffic intersection vehicle queue trajectory efficient planning system, characterized in that, Including: A scenario construction module configured to construct a control area based on signal timing information, the initial speeds and initial times of connected and automated vehicles and human-driven vehicles entering the control area; A candidate trajectory construction module configured to construct an initial segmented trajectory based on passing through the intersection at the highest speed limit and adjust the trajectory according to safety boundary constraints to obtain a candidate segmented trajectory; The final trajectory construction module is configured to, if the candidate segmented trajectory arrives at the intersection during the red light period, based on the start time of the next green light: for the connected and autonomous vehicle, construct a uniformly accelerated section in reverse and connect it to the candidate segmented trajectory through a connection section to obtain the final segmented trajectory; for the human-driven vehicle, construct a parking section in reverse and connect it to the candidate segmented trajectory through a connection section to obtain the final segmented trajectory; the connection section includes a uniformly decelerated section, or a combination of a uniformly decelerated section and a parking section; For the human-driven vehicle, dynamically adjust the accelerations of its uniformly accelerated section and uniformly decelerated section to adapt to the trajectory estimation error of the human-driven vehicle in the mixed traffic flow, specifically expressed as: ； Wherein, and represent the maximum acceleration and the maximum deceleration, is a coefficient; In a CAV-HV mixed traffic environment, the trajectory of a CAV may be affected by the HV in front. Since the HV arriving at the intersection during the red light needs to stop and wait, and starts to accelerate through the intersection from a stationary state after the green light turns on; when the rear CAV arrives at the intersection at its maximum speed, the HV may not have accelerated to its maximum speed yet. At this time, if the distance between the two vehicles is already relatively close, the CAV must decelerate to maintain a safe distance from the HV in front, which will disrupt the stability of the traffic flow behind the CAV and requires delaying the arrival time of the CAV at the intersection by a certain amount of time , so that the trajectory of the CAV can smoothly merge into its safety boundary , and the delayed arrival time is called the expected arrival time, denoted as , Among them, is the delay time for the CAV to reach the intersection, is the maximum speed, is the trajectory on the speed at a point, is the acceleration of the HV trajectory; is the vehicle perception and reaction time, is the minimum distance after the vehicle makes an emergency brake, is the desired arrival time, and L is the length of the control area; The trajectory update module is configured to store the final segmented trajectory of the current vehicle in a queue, and when a subsequent vehicle enters the control area, iteratively update the trajectory information to achieve trajectory planning.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by a processor, it implements the steps in a method for efficient trajectory planning of a vehicle queue at an intersection with connected and autonomous vehicle and human-driven vehicle mixing as described in any one of claims 1-6.

9. A computer device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps in a method for efficient trajectory planning of a vehicle queue at an intersection with connected and autonomous vehicle and human-driven vehicle mixing as described in any one of claims 1-6.

Images