Multi-lane cooperative lane changing control method based on expressway interchange area
By constructing a three-dimensional spatiotemporal graph and a vehicle kinematic model, and combining the convex feasible set algorithm to generate the optimal lane-changing trajectory, the traffic conflict problem in multi-lane weaving areas is solved, improving traffic efficiency and resource utilization.
Patent Information
- Application Number
- CN202511545661.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-10-28
- Publication Date
- 2026-01-23
AI Technical Summary
Existing technologies struggle to achieve multi-lane coordinated lane-changing control in urban expressway interchanges, leading to increased traffic conflicts and reduced traffic efficiency. This is especially true under conditions of high traffic load and high lane-changing ratio, where existing methods suffer from low computational efficiency and a singular optimization objective.
By constructing a three-dimensional spatiotemporal map, using intelligent connected vehicle technology and vehicle-road cooperative technology, we can obtain vehicle information in all time and space. Combining vehicle kinematics model and convex feasible set algorithm, we can seek the set of feasible lane-changing channels for the target vehicle, and generate the optimal lane-changing trajectory through channel selection objective function and speed smoothing objective function.
It improves traffic efficiency and spatial-temporal resource utilization in multi-lane weaving areas, maintaining good control, especially under conditions of high traffic load and high lane-changing ratio.
Smart Images
Figure CN121393142A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of intelligent connected vehicle cooperative control technology, specifically relating to a multi-lane cooperative lane-changing control method based on the interchange weaving zone of expressways. Background Technology
[0002] Interchanges on urban expressways are critical bottlenecks in traffic flow. Frequent merging and lane-changing maneuvers in these areas exacerbate traffic conflicts and reduce efficiency. In an intelligent connected vehicle environment, using vehicle-to-infrastructure (V2I) technology to achieve coordinated lane-changing control is an effective way to improve the efficiency and safety of these interchanges.
[0003] Existing technologies mostly focus on the coordinated control of adjacent lanes on main and auxiliary roads, making it difficult to adapt to complex multi-lane traffic scenarios. Under high traffic loads, some methods tend to concentrate lane-changing points upstream of the control area, resulting in underutilization of mid- and downstream spatiotemporal resources and limited traffic potential. Furthermore, existing trajectory planning methods often face challenges such as low computational efficiency and a singular optimization objective when dealing with multi-vehicle, multi-lane cooperative problems.
[0004] To overcome the above shortcomings, this invention proposes a multi-lane collaborative lane-changing control method based on the interchange weaving area of expressways. By constructing a three-dimensional spatiotemporal map and feasible lane-changing channels, it achieves efficient and safe collaborative control of multi-lane intelligent connected vehicles, optimizes the distribution of lane-changing points, and improves overall traffic efficiency. Summary of the Invention
[0005] The purpose of this invention is to provide a multi-lane cooperative lane-changing control method based on the weaving area of expressway interchanges, which overcomes the shortcomings of the prior art, can effectively improve the traffic efficiency and spatial-temporal resource utilization rate of the multi-lane weaving area, and can still maintain good control effect under high traffic load and high lane-changing ratio conditions.
[0006] To solve the above problems, the technical solution adopted by the present invention is as follows: A method for multi-lane cooperative lane-changing control in expressway interchange weaving areas includes the following steps: Step 1: Using intelligent connected vehicle technology and vehicle-road cooperative technology, obtain the trajectory and status information of all vehicles in the entire time and space range of the control area of the expressway interchange weaving zone. By introducing a time axis on the basis of a two-dimensional environmental map, integrate the above vehicle information to construct a three-dimensional spatiotemporal map of the control area. Step 2: Within the three-dimensional spatiotemporal graph constructed in Step 1, a simplified two-degree-of-freedom bicycle model is used as the vehicle kinematics model. The dynamic obstacles in the control area are processed by combining the convex feasible set algorithm. By setting a vehicle speed cyclic adjustment mechanism, the feasible lane-changing channel set of the target vehicle is sought to solve the non-convex optimization lane-changing channel search problem in a multi-lane dynamic environment. Step 3: Based on the feasible lane-changing channel set obtained in Step 2, establish a dual-objective optimization model of channel selection objective function and speed smoothing. The channel selection objective function realizes the priority ranking of lane-changing channels by quantifying efficiency cost and safety cost. The speed smoothing objective function selects the optimal lane-changing point with the goal of minimizing speed fluctuation before and after lane changing. The lane-changing trajectory of the target vehicle is generated by a fifth-order polynomial trajectory function.
[0007] Furthermore, the specific method for constructing the three-dimensional spatiotemporal map in step 1 is as follows: Based on the two-dimensional environmental map, a time axis is introduced. Based on the multi-lane collaborative lane-changing control framework of the urban expressway interchange, the control method is activated when the target vehicle touches the boundary of the control area. High-precision perception is achieved by using intelligent connected vehicle technology and vehicle-road cooperative technology. After obtaining vehicle information in all time and space, the dynamic time step is calculated according to the real-time traffic load. The vehicle information and environmental information are integrated to construct a three-dimensional spatiotemporal map, wherein the traffic load is calculated in real time by the vehicle traffic volume per unit time within the control area.
[0008] Furthermore, the feasible lane-changing channel search method described in step 2 specifically includes the following steps: Step 21: Construct a two-degree-of-freedom bicycle model. The velocity of the bicycle at the center of the rear axle is: ; Front and rear axle kinematic constraints are ; Based on the geometric relationship between the front and rear wheels: ; The yaw rate is: ; The turning radius is: ; The front wheel deflection angle is: ; When the state variable is The control quantity is Then, the vehicle kinematic model can be obtained as follows: ; In intelligent connected vehicle control, the general controlled object, the vehicle kinematic model, can be transformed into: ; in, The speed of the vehicle at the center of the rear axle. Wheelbase The instantaneous turning radius of the rear axle center. This refers to the front wheel deflection angle; Step 22: Transform the trajectory planning problem into a convex optimization problem, construct a convex set, and define the collision avoidance constraints between the vehicle and obstacles: ; In the formula, For obstacle space; The distance in Cartesian coordinates; Minimum safe distance; The non-convex obstacle avoidance constraint is made convex and then solved iteratively using the convex feasible set algorithm. The iteration termination condition is: ; Step 23: Apply speed control to the vehicle. Constraints are set, including the initial velocity range. : in, The desired speed of the target vehicle, This represents the number of loop iterations. The velocity parameters are dynamically adjusted based on the solution results of the convex feasible set algorithm. Continue until a feasible set of lane-changing channels is obtained.
[0009] Furthermore, the convex optimization objective function of the convex feasible set algorithm described in step 22 is: , Simultaneously satisfying the speed constraint Acceleration constraints and convex obstacle avoidance constraints ; in, The weights of the objective function, For the difference operator between velocity and acceleration, .
[0010] Furthermore, the objective function for channel selection in step 3 is: , in, For efficiency and cost, For safety costs; For parameters.
[0011] Furthermore, the security cost is: , Among them, the degree of danger when the two vehicles are in the same direction and at the same speed. , ; in, It is the vehicle's deceleration; It refers to the vehicle's reaction time; It is the safe following distance between vehicles; It is a relative distance; Danger level when two vehicles are in the same direction but traveling at different speeds ,
[0012] in, The yaw angle deviation between the target vehicle and surrounding vehicles. Used to determine whether the speeds of two vehicles are in the same direction or opposite directions.
[0013] Furthermore, the velocity smoothing objective function described in step 3 is:
[0014] In the formula: These represent the average speeds of the vehicles traveling from the starting point to the lane change point and from the lane change point to the destination, respectively.
[0015] Furthermore, the fifth-order polynomial lane-changing trajectory function mentioned in step 3 is:
[0016] in, The displacement of the trolley is a fifth-order polynomial in terms of time. Taking its first and second derivatives yields the velocity and acceleration state curves. Position state curve:
[0017] Velocity state curve:
[0018] Acceleration state curve:
[0019] The state curve changes over time. Let... This is the initial moment for the car. At the final moment, combining the state equations for position, velocity, and acceleration, we construct them into a matrix form, where... These represent longitudinal and lateral movements, respectively.
[0020] .
[0021] Compared with the prior art, the present invention has the following beneficial effects: The multi-lane cooperative lane-changing control method based on the expressway interchange weaving area described in this invention utilizes intelligent connected vehicle technology and vehicle-road cooperative technology to construct a three-dimensional spatiotemporal map of the control area. Based on the vehicle kinematic model, it seeks the set of feasible lane-changing channels for the target vehicle, selects the optimal lane-changing channel and the optimal lane-changing point, and generates the lane-changing trajectory of the target vehicle. This method can effectively improve the traffic efficiency and spatiotemporal resource utilization rate of the multi-lane weaving area, and can still maintain good control performance under high traffic load and high lane-changing ratio conditions. Attached Figure Description
[0022] Figure 1 This is a framework diagram of the coordinated lane-changing control in the weaving area of a multi-lane coordinated lane-changing control method based on the weaving area of an expressway interchange.
[0023] Figure 2 This is a three-dimensional spatiotemporal diagram of the control area in the multi-lane cooperative lane-changing control method based on the interchange weaving area of expressways.
[0024] Figure 3 This is a schematic diagram of the lane-changing trajectory in the multi-lane cooperative lane-changing control method based on the weaving area of the expressway interchange.
[0025] Figure 4 This is a heat map showing the distribution of multi-lane lane-changing points in a multi-lane cooperative lane-changing control method based on the interchange weaving area of expressways.
[0026] Figure 5 This is a heatmap of the spatiotemporal utilization of multiple lanes in the multi-lane cooperative lane-changing control method based on the interchange weaving area of expressways. Detailed Implementation
[0027] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0028] In summary, the multi-lane cooperative lane-changing control method based on the weaving area of an expressway interchange as described in this invention includes the following steps: The first step, in the context of intelligent connected and autonomous driving, is the coordinated control framework for the interleaving zone, as follows: Figure 1 As shown, the upper section is the main road and the lower section is the auxiliary road, which are interconnected. The length of the interconnected area is 120m.
[0029] The collaborative lane-changing control in the interchange will take the roadside auxiliary computing unit as the control center, and rely on intelligent roadside information collection equipment and vehicle equipment to obtain accurate all-time and all-space traffic information in the collaborative control area. After processing and collaborative decision-making by the control center, a precise control scheme for intelligent connected vehicles will be formed.
[0030] The second step involves a multi-lane cooperative lane-changing control framework based on urban expressway interchange weaving areas. When a target vehicle touches the boundary of the control area, the multi-lane cooperative lane-changing control method is activated. Leveraging intelligent connected vehicle technology and vehicle-road cooperative technology, the target vehicle can achieve high-precision perception of the road traffic environment, acquiring trajectory and status information of all vehicles within the control area across all time and space. By integrating all vehicle information within the control area, a time axis is introduced on top of a two-dimensional environmental map to construct a three-dimensional spatiotemporal map of the control area, as shown below. Figure 2 As shown.
[0031] The third step is to construct a vehicle kinematics model within the three-dimensional spatiotemporal map and apply the convex feasible set algorithm to the obstacles to find the set of feasible lane-changing channels for the target vehicle.
[0032] (1) Construct a vehicle kinematic model, considering the vehicle's kinematic characteristics and the constraints imposed by surrounding vehicles and dynamic obstacles, and describe the vehicle's motion state from a geometric perspective. The following basic assumptions need to be made in the vehicle kinematic model: ① Assume the vehicle is a rigid body, and its shape and size are fixed; ② Assume the vehicle moves on a two-dimensional plane, and do not consider the vehicle's motion in the Z-axis direction; ③ Assuming the vehicle's center of gravity does not move, that is, the contact point between the wheels and the road surface does not slip during the vehicle's movement, thus reducing the impact of friction on the vehicle. ④ It is assumed that the left and right tires of the vehicle have the same steering angle and speed at all times, so as to combine the left and right tires of the vehicle into a single tire description. ⑤ Assume that the aerodynamic characteristics of the vehicle can be ignored, that is, the air resistance and lift experienced by the vehicle during its movement can be ignored.
[0033] Based on the above assumptions, the kinematic model calculation is simple, and the vehicle kinematic model can be simplified to a two-degree-of-freedom bicycle model.
[0034] (2) The convex feasible set algorithm can take into account the kinematic characteristics of the vehicle and the surrounding environment obstacles during the planning process, transforming the trajectory planning problem into a convex optimization problem. It constructs a safe passage in the local convex space to avoid collisions with other vehicles, thereby obtaining a set of feasible trajectories of the target vehicle. It is suitable for intelligent connected vehicles to quickly find the optimal motion trajectory in complex environments.
[0035] Because vehicles within the control area are constantly in motion, and the information about obstacles surrounding the target vehicle changes continuously at each time step, the optimization problem exhibits non-convexity. Traditional non-convex optimization methods have limitations; therefore, the convex feasible set algorithm is applied.
[0036] Specifically, the following steps are included: First, we need to construct a convex set. A set is convex if any line segment connecting any two points within the set is also within the set. Assume the state of a vehicle is defined as... The vehicle from Move to Its trajectory can be represented as in For state space, For vehicles in The state at any given moment, Within the planned area, the vehicle status is as follows. When, the area occupied in Cartesian space is ,in Let be the dimension of the Cartesian space.
[0037] The distance between a vehicle and an obstacle can be represented by Euclidean distance. To avoid a collision between the vehicle and the obstacle, the following constraints must be met:
[0038] in, For obstacle space; The distance in Cartesian coordinates; This is the minimum safe distance.
[0039] The constraint set obtained from the above formula is non-convex, making it difficult to solve the planning problem. Therefore, it is necessary to process the obstacle constraints into convex ones.
[0040] Assuming obstacle avoidance constraints It can be decomposed into several subsets, and a subset can be represented as ,in It is about A smooth convex function has a positive definite matrix. , making The convex form of the constraint is defined as follows: ,if If it is a convex set, then ;if If the complement of is a convex set, then it is . Convex function, i.e. ,in, This serves as a reference point for vehicle planning. A convex constraint can be expressed as:
[0041] The convex optimization problem can be represented by minimizing the error between the target trajectory and the reference trajectory:
[0042] in, The weights of the objective function, For the difference operator between velocity and acceleration, .
[0043] The solution to convex optimization problems can be achieved using the convex feasible set algorithm. The solution process is as follows:
[0044] in, It is a convex feasible set.
[0045] The convex feasible set algorithm iterates continuously, using the optimal solution obtained as a reference point for the next iteration. There are two conditions for termination: iteration terminates when the solutions from two iterations are small or the cost function values from two iterations are small, as shown in the following equation:
[0046] (3) Based on the vehicle kinematics model, a cycle is set for the vehicle speed. By adjusting the speed of the target vehicle, the convex feasible set algorithm is applied to the obstacle to find the set of feasible lane-changing channels for the target vehicle.
[0047] Specifically, the following steps are included: First, set the initial speed of the target vehicle to be... .in , where is the desired speed of the target vehicle, and is the number of cycles. The target vehicle moves at a constant speed until it reaches the desired speed when there are no obstacles ahead; if there are obstacles ahead, the vehicle decelerates uniformly until it reaches the same speed as the vehicle in front.
[0048] If the target vehicle is traveling at the generated initial speed and the convex feasible set algorithm fails to find a solution, meaning no feasible lane-changing route set can be found, then speed adjustment is performed. The process continues until the adjusted speed allows the convex feasible set algorithm to find the optimal solution, thus obtaining a feasible set of lane-changing channels.
[0049] The fourth step involves establishing a channel selection objective function and a speed smoothing objective function for the lane-changing trajectory generation process. The optimal lane-changing channel and the optimal lane-changing point are selected through the objective functions, and then the lane-changing trajectory of the target vehicle is generated through a fifth-order polynomial lane-changing trajectory function.
[0050] The lane-changing route is selected by weighing the costs of all feasible lane-changing routes in the feasible lane-changing route set, and then selecting the route with the highest cost function value as the lane-changing route for the target vehicle.
[0051] Specifically, the following steps are included: First, efficiency cost is defined as the average speed at which the target vehicle travels within the lane-changing lane.
[0052]
[0053] in, For channel duration, The speed of the target vehicle at each time point in the corridor.
[0054] Next, the safety cost is defined as the degree of danger posed to the target vehicle by the vehicle with the highest risk within the lane-changing lane. A risk assessment function is established based on the safe distance between the target vehicle and surrounding vehicles.
[0055]
[0056] in, It is a positive linear function; only the positive part is retained.
[0057] When there are multiple vehicles in the vicinity, the safety cost is determined by the level of danger posed by the vehicle with the highest risk to the target vehicle.
[0058]
[0059] The degree of danger between two vehicles can be solved using a safe distance model, mainly divided into two cases: Case 1: When the positions and speeds of the two vehicles are in the same direction, the degree of danger can be determined by the safe distance. and relative distance The ratio is expressed as .
[0060]
[0061] in, It is the vehicle's deceleration; It refers to the vehicle's reaction time; It refers to the safe following distance between vehicles.
[0062] Scenario 2: When two vehicles are positioned in the same direction but traveling at different speeds. Since collisions primarily occur longitudinally, the degree of danger posed by surrounding vehicles to the target vehicle is mainly determined by the target vehicle's speed. Surrounding vehicles The components of direction.
[0063]
[0064] in, The yaw angle deviation between the target vehicle and surrounding vehicles. Used to determine whether the speeds of two vehicles are in the same direction or opposite directions.
[0065] Finally, the objective function for channel selection is as follows:
[0066] in, For efficiency and cost, For safety costs; For parameters.
[0067] By calculating and normalizing the efficiency and safety costs of the target vehicle, the objective function value for lane selection is obtained. The lane-changing lane with the larger objective function value is selected as the optimal solution. At that time, the largest right boundary within the channel will be... Compare and select The largest lane is used as the lane-changing lane for the target vehicle.
[0068] (1) Using the speed smoothing objective function, by measuring the cost function of each lane-changing point in the lane-changing channel, the point with the smallest speed fluctuation and the smoothest trajectory is selected as the lane-changing point of the target vehicle. With the goal of minimizing the speed smoothing objective function value, the point with relatively stable speed before and after is selected as the lane-changing point of the target vehicle, so as to reduce the speed fluctuation of the target vehicle during operation.
[0069]
[0070] In the formula: These represent the average speeds of the vehicles traveling from the starting point to the lane change point and from the lane change point to the destination, respectively.
[0071] (2) A fifth-order polynomial lane-changing trajectory function is used to generate the lane-changing trajectory. A reference coordinate system is established with the initial position of the target vehicle during lane changing, with the vehicle's center of gravity as the origin and the vehicle's direction of travel as the coordinate system. The axis is based on the direction of vehicle lane changing. Axis, based on the time after the track change begins Establish a vehicle coordinate system for the variables, such as Figure 3 As shown.
[0072] Specifically, the following steps are included: The lane-changing trajectory function of the target vehicle can be expressed as:
[0073] in, The displacement of the car is relative to time. The velocity and acceleration state curves are obtained by differentiating the fifth-order polynomial with respect to its first and second orders.
[0074] Position state curve:
[0075] Velocity state curve:
[0076] Acceleration state curve:
[0077] The state curve changes over time. Let... This is the initial moment for the car. At the final moment, combining the state equations for position, velocity, and acceleration, we construct them into a matrix form, where... These represent longitudinal and lateral movements, respectively.
[0078] The fifth step involved constructing a multi-lane traffic simulation model to explore the impact of different lane-changing ratios on multi-lane traffic in urban expressway weaving areas under differentiated traffic loads.
[0079] (1) Construct a multi-lane traffic simulation model, including the configuration of the simulation environment; Specifically, the following steps are included: Based on actual road conditions, a 150m long road segment containing two main roads and one auxiliary road was created, with a simulation duration of 1 hour. Vehicles were generated according to a Poisson distribution. The desired speed of the vehicles was set to 55km / h, the maximum speed to 60km / h, and the maximum acceleration to 3m / s². 2 The following behavior of intelligent connected vehicles is simulated using the CACC (Cooperative Adaptive Cruise Control) model.
[0080] For intelligent connected vehicles, a lane-changing model is constructed to achieve cooperative lane-changing control between multiple lanes. Specifically, the lane-changing model in this method implements lane-changing decisions by constructing a lane-changing channel search cost function, the mathematical expression of which is:
[0081] (2) Using traffic load and lane change ratio as independent variables, simulate different traffic loads and different lane change ratios to explore the impact on multiple lanes in the urban expressway weaving area. Compare the distribution of lane change points and the spatiotemporal utilization rate of the urban expressway weaving area under the collaborative lane change control method based on the lane change channel search cost function under different environments, and analyze the control effect of the method.
[0082] Based on traffic flow data collected on-site from expressway interchange weaving areas, traffic loads are divided into five groups: 400, 800, 1200, 1600, and 2000 (pcu / h), with lane-changing ratios k set at 0.30, 0.32, 0.34, 0.36, and 0.40, respectively.
[0083] Under different traffic loads and lane-changing ratios, the distribution of lane-changing points in multi-lane traffic exhibits different characteristics. A heatmap of the distribution of lane-changing points in multi-lane traffic is shown below. Figure 4As shown, the horizontal axis represents the location of the vehicle in the control area. The first row of the heat map shows the distribution of lane change points from Main Road 2 to Main Road 1, the second row shows the distribution of lane change points from Main Road 1 to the auxiliary road, and the third row shows the distribution of lane change points from the auxiliary road to Main Road 1.
[0084] The following can be observed from the diagram: ① The distribution patterns of lane-change points for the three lanes are basically consistent. Under the same lane-change ratio, as traffic load increases, the distribution of lane-change points gradually shifts from the downstream to the upstream of the control area; under the same traffic load, as the lane-change ratio increases, the distribution of lane-change points also gradually shifts upstream of the control area. ② The distribution patterns of lane-change points between adjacent lanes on the main and auxiliary roads are similar. Compared to adjacent lanes on the main and auxiliary roads, the lane-change point from Main Road 2 to Main Road 1 is closer to the upstream of the control area, indicating a more urgent lane-change need.
[0085] The time-space utilization rate is defined as the ratio of the time a vehicle spends on the road to the total length of the statistical period. This indicator can reflect the occupancy of road space.
[0086] Based on the distribution of multi-lane lane changing points, draw... Figure 5 The heatmaps shown depict the spatial and temporal utilization of multiple lanes (Figure a represents main road 2, Figure b represents main road 1, and Figure c represents auxiliary roads). The horizontal axis indicates the location of vehicles in the weaving zone, and the vertical axis shows the lane-changing ratio gradually increasing from bottom to top. The color intensity of the blocks in the heatmap represents the spatial and temporal utilization rate of that area. Figure 5 It can be seen that: ① The spatiotemporal utilization rates of adjacent lanes on the main and auxiliary roads are basically consistent across different sections of the weaving zone and coordinated control area. As traffic load and lane-changing ratio increase, the spatiotemporal utilization rate of the upstream section within the control area gradually increases. ② Under the same lane-changing ratio and traffic load conditions, the spatiotemporal utilization rate of the auxiliary road is closer to the upstream of the control area compared to main road 1. ③ The spatiotemporal utilization rate of main road 2 is mainly concentrated in the middle and upper reaches of the weaving zone, gradually approaching the upstream of the weaving zone as traffic load and lane-changing ratio increase.
[0087] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.
Claims
1. A method for multi-lane cooperative lane-changing control in expressway interchange weaving zones, characterized in that, Includes the following steps: Step 1: Using intelligent connected vehicle technology and vehicle-road cooperative technology, obtain the trajectory and status information of all vehicles in the entire time and space range of the control area of the expressway interchange weaving zone. By introducing a time axis on the basis of a two-dimensional environmental map, integrate the above vehicle information to construct a three-dimensional spatiotemporal map of the control area. Step 2: Within the three-dimensional spatiotemporal graph constructed in Step 1, a simplified two-degree-of-freedom bicycle model is used as the vehicle kinematics model. The dynamic obstacles in the control area are processed by combining the convex feasible set algorithm. By setting a vehicle speed cyclic adjustment mechanism, the feasible lane-changing channel set of the target vehicle is sought to solve the non-convex optimization lane-changing channel search problem in a multi-lane dynamic environment. Step 3: Based on the feasible lane-changing channel set obtained in Step 2, establish a dual-objective optimization model of channel selection objective function and speed smoothing. The channel selection objective function realizes the priority ranking of lane-changing channels by quantifying efficiency cost and safety cost. The speed smoothing objective function selects the optimal lane-changing point with the goal of minimizing speed fluctuation before and after lane changing. The lane-changing trajectory of the target vehicle is generated by a fifth-order polynomial trajectory function.
2. The method for multi-lane cooperative lane-changing control in a rapid transit interchange weaving zone according to claim 1, characterized in that, The specific method for constructing the three-dimensional spatiotemporal map in step 1 is as follows: Based on the two-dimensional environmental map, a time axis is introduced. Based on the multi-lane collaborative lane-changing control framework of the urban expressway interchange, the control method is activated when the target vehicle touches the boundary of the control area. High-precision perception is achieved by using intelligent connected vehicle technology and vehicle-road cooperative technology. After obtaining vehicle information in all time and space, the dynamic time step is calculated according to the real-time traffic load. The vehicle information and environmental information are integrated to construct a three-dimensional spatiotemporal map, where the traffic load is calculated in real time by the vehicle traffic volume per unit time within the control area.
3. The method for multi-lane cooperative lane-changing control in a rapid transit interchange weaving zone according to claim 1, characterized in that, The feasible lane-changing channel search method described in step 2 specifically includes the following steps: Step 21: Construct a two-degree-of-freedom bicycle model. The speed of the bicycle at the center of the rear axle is... ; Front and rear axle kinematic constraints are ; Based on the geometric relationship between the front and rear wheels: ; The yaw rate is: ; The turning radius is: ; The front wheel deflection angle is: ; When the state variable is The control quantity is Then, the vehicle kinematic model can be obtained as follows: ； In intelligent connected vehicle control, the general controlled object The vehicle kinematics model can be transformed into: ; in, The speed of the vehicle at the center of the rear axle. Wheelbase The instantaneous turning radius of the rear axle center. This refers to the front wheel deflection angle; Step 22: Transform the trajectory planning problem into a convex optimization problem, construct a convex set, and define the collision avoidance constraints between the vehicle and obstacles. ; In the formula, For obstacle space; The distance in Cartesian coordinates; Minimum safe distance; The non-convex obstacle avoidance constraints are made convex, and then iteratively solved using the convex feasible set algorithm. The iteration termination condition is: ； Step 23: Apply speed control to the vehicle. Constraints are set, including the initial velocity range. ； in, The desired speed of the target vehicle, This represents the number of loop iterations. The velocity parameters are dynamically adjusted based on the solution results of the convex feasible set algorithm. Continue until a feasible set of lane-changing channels is obtained.
4. The method for multi-lane cooperative lane-changing control in a rapid transit interchange weaving zone according to claim 3, characterized in that, The convex optimization objective function of the convex feasible set algorithm described in step 22 is: ， Simultaneously satisfying the speed constraint Acceleration constraints and convex obstacle avoidance constraints ; in, The weights of the objective function, For the difference operator between velocity and acceleration, .
5. The method for multi-lane cooperative lane-changing control in expressway interchange weaving areas according to claim 1, characterized in that, The objective function for channel selection in step 3 is: ； in, For efficiency and cost, For safety costs; For parameters.
6. The method for multi-lane cooperative lane-changing control in the weaving area of an expressway interchange according to claim 5, characterized in that, The security cost is: ， Among them, the degree of danger when the two vehicles are in the same direction and at the same speed. , ； in, It is the vehicle's deceleration; It refers to the vehicle's reaction time; It is the safe following distance between vehicles; It is a relative distance; Danger level when two vehicles are in the same direction but traveling at different speeds , ； in, The yaw angle deviation between the target vehicle and surrounding vehicles. Used to determine whether the speeds of two vehicles are in the same direction or opposite directions.
7. The method for multi-lane cooperative lane-changing control in the weaving zone of an expressway interchange according to claim 1, characterized in that, The velocity smoothing objective function mentioned in step 3 is: ； In the formula: These represent the average speeds of the vehicles traveling from the starting point to the lane change point and from the lane change point to the destination, respectively.
8. The method for multi-lane cooperative lane-changing control in expressway interchange weaving areas according to claim 1, characterized in that, The fifth-order polynomial lane-changing trajectory function mentioned in step 3 is: ； in, The displacement of the trolley is a fifth-order polynomial in terms of time. Taking its first and second derivatives yields the velocity and acceleration state curves. Position state curve: ; Velocity state curve: ; Acceleration state curve: ; The state curve changes over time. Let... This is the initial moment for the car. At the final moment, combining the state equations for position, velocity, and acceleration, we construct them into a matrix form, where... These represent longitudinal and lateral movements, respectively. ； 。