Layered cooperative confluence control method for mixed traffic flow

Abstract

The invention discloses a mixed traffic flow hierarchical cooperative confluence control method, which belongs to the field of automatic driving, and comprises the following steps: constructing a utility function according to mixed traffic flow vehicle interaction characteristics, and constructing a confluence sequence optimization integer programming model according to the utility function; a motorcade communication network is constructed according to the vehicle real-time state information, a vehicle distance constraint and vehicle dynamics model is constructed according to the motorcade communication network, and a consistency controller is constructed according to the vehicle dynamics model; the method comprises the steps of obtaining vehicle real-time information, optimizing the vehicle real-time information through a confluence sequence optimization integer programming model to obtain an optimal confluence sequence, and performing calculation through a consistency controller according to the optimal confluence sequence and the vehicle real-time information to obtain vehicle control information. The interaction level of an automatic driving vehicle and a manual driving vehicle can be improved, and the automatic driving safety is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of autonomous driving, and particularly relates to a hierarchical collaborative merging control method for mixed traffic flow. Background Art

[0002] There are a large number of complex interactions between ramp vehicles and mainline vehicles in the highway merging area, which thus becomes a traffic bottleneck. With the development of vehicle-to-everything (V2X) and autonomous vehicle technologies, strategies for realizing collaborative merging from micro-trajectory control are expected to be achieved. However, due to the slow growth of the current penetration rate of autonomous vehicles and the existence of a large number of human-driven vehicles, a pure autonomous driving environment still takes a long time to be realized. Different from autonomous vehicles, human-driven vehicles cannot be directly controlled and exhibit a high degree of randomness and uncertainty. Therefore, developing a merging control strategy for mixed traffic flow is still an important challenge at present.

[0003] To solve the above problems, many collaborative merging strategies have been proposed in existing research, but there are still many difficulties in practical applications. First of all, most of the existing research focuses on a pure autonomous driving environment and does not fully consider the behavioral uncertainty of human-driven vehicles. Secondly, the existing research adopts a merging sorting based on preset rules or logic, which cannot guarantee the optimal merging order and cannot accurately capture the interaction between vehicles. Finally, the existing controllers assume that all following vehicles in the platoon will adjust their speeds to match the leading vehicle to achieve a consistent state, which is more applicable to a pure autonomous driving environment. However, in a mixed traffic environment, due to the uncertainty of human driving behavior, the existing controllers are difficult to ensure the stability of the platoon and may cause potential safety problems. These deficiencies reduce the interaction level between autonomous vehicles and human-driven vehicles and limit the application of autonomous driving technology in a mixed traffic flow environment. Summary of the Invention

[0004] To solve the above technical problems, the present invention proposes a hierarchical collaborative merging control method for mixed traffic flow to solve the problems existing in the above prior art.

[0005] To achieve the above object, the present invention provides a hierarchical collaborative merging control method for mixed traffic flow, including:

[0006] Construct a utility function according to the interaction characteristics of mixed traffic flow vehicles, and construct an integer programming model for optimizing the merging order according to the utility function;

[0007] Construct a platoon communication network according to the real-time state information of vehicles, construct a headway strategy and a vehicle dynamics model according to the platoon communication network, and construct a consensus controller according to the vehicle dynamics model; wherein the gain coefficient in the consensus controller is constrained by local stability and queue linear constraint stability analysis.

[0008] Obtain the real-time information of the vehicle, optimize the real-time information of the vehicle through the merging sequence optimization integer programming model to obtain the optimal merging sequence, and according to the optimal merging sequence, calculate based on the real-time information of the vehicle through the consistency controller to obtain the vehicle control information.

[0009] Optionally, the utility function is:

[0010]

[0011] Wherein, represents the traffic efficiency index, represents the driving comfort index, represents the traffic safety index, 、 、 represent the weight factors, represents the total cost.

[0012] Optionally, the acquisition process of different indexes in the utility function includes:

[0013]

[0014]

[0015]

[0016] Wherein, 、 and represent the longitudinal position, speed and acceleration of vehicle , represents the desired speed, represents the safe vehicle spacing, represents the time interval, represents the initial time, , represents the vehicle label, represents all vehicles.

[0017] Optionally, the merging sequence optimization integer programming model is:

[0018]

[0019]

[0020] Wherein, represents the total cost of the given merging sequence, represents the decision variable, indicating whether to select edge , if edge is selected, then , otherwise .

[0021] Optionally, the construction process of the vehicle fleet communication network includes:

[0022] According to the real-time vehicle status information, a vehicle fleet communication network is constructed, where the vehicle fleet communication network is a directed graph structure, including a node set, an edge set, and a weighted adjacency matrix. The node set includes the control vehicle in the vehicle fleet. The edge set includes the relationships between vehicles. The weighted adjacency matrix indicates whether there is communication between vehicles;

[0023] The vehicle fleet communication network is described by a Laplacian matrix.

[0024] Optionally, the vehicle spacing strategy is:

[0025]

[0026]

[0027] Among them, represents the desired constant headway time, represents vehicle at time longitudinal speed, represents vehicle length, represents the minimum spacing between vehicles, represents the vehicle label, represents the node set, represents vehicle desired spacing from the leading vehicle in the vehicle fleet, represents vehicle desired headway time from the leading vehicle in the vehicle fleet.

[0028] Optionally, the vehicle dynamics model is:

[0029]

[0030] Among them, represents the jerk, represents the acceleration, represents the desired acceleration generated by the consensus controller, represents vehicle time delay to achieve the acceleration, represents the ratio of the achievable desired acceleration, represents the vehicle label.

[0031] Optionally, the consensus controller includes:

[0032] Control the longitudinal position, longitudinal speed, and acceleration of the vehicle, where the error between the longitudinal position and the desired position, the error between the longitudinal speed and the desired speed, and the acceleration gradually approach zero.

[0033] Optionally, the process of constraining the gain coefficient in the consensus controller includes:

[0034] Conduct local stability analysis using the Lyapunov-Razumikhin theorem to obtain local gain coefficient constraints;

[0035] Conduct platoon linear stability analysis using the final platoon linear stability theorem to obtain platoon linear gain coefficient constraints;

[0036] Constrain the gain coefficient by obtaining local gain coefficient constraints and platoon linear gain coefficient constraints.

[0037] Compared with the prior art, the present invention has the following advantages and technical effects:

[0038] The present invention provides a non-centralized vehicle cooperative merging motion planning and control framework to improve regional traffic efficiency and safety by controlling autonomous vehicles. First, a merging sorting optimization model based on integer programming is proposed, a human-like interactive decision-making model is established, and at the same time, the merging sorting problem is transformed into a shortest path problem to reduce the computational cost; second, a consensus controller with communication delay is proposed, using the error information of all vehicles in front of the ego vehicle as the control input, considering the dynamic acceleration change of the leading vehicle, and enhancing the stability and robustness of the mixed-flow platoon; finally, the stability conditions of the mixed-flow platoon are analyzed, including local stability and platoon linear stability, and a method for setting the gain coefficient in the control model is provided.

[0039] The cooperative merging control method for mixed traffic flow proposed by the present invention establishes a hierarchical optimization model, considers the uncertainty of the behavior of human-driven vehicles, and stably and efficiently solves the optimal control of autonomous vehicles; improves the traffic efficiency and safety in the merging area by optimizing the merging order; establishes a consensus controller, and based on the optimal merging order, realizes the cooperative merging longitudinal acceleration control of vehicles; gives a method for determining the gain coefficient in the model, and enhances the stability and robustness of the mixed-flow platoon. Finally, it realizes good interaction between autonomous vehicles and human-driven vehicles, and improves the traffic operation efficiency and safety in the merging area. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] The drawings constituting a part of this application are used to provide a further understanding of this application. The schematic embodiments of this application and their descriptions are used to explain this application and do not constitute an improper limitation of this application. In the drawings:

[0041] Figure 1It is the flowchart of the method according to the embodiment of the present invention;

[0042] Figure 2 It is the schematic diagram of the feasible and infeasible orders of the confluence sorting according to the embodiment of the present invention;

[0043] Figure 3 It is the schematic diagram of the feasible region of the controller gain parameters according to the embodiment of the present invention;

[0044] Figure 4 It is the effect diagram of the consistency controller according to the embodiment of the present invention. Specific embodiments

[0045] It should be noted that, without conflict, the embodiments in the present application and the features in the embodiments can be combined with each other. The present application will be described in detail below with reference to the drawings and in conjunction with the embodiments.

[0046] It should be noted that the steps shown in the flowchart of the drawings can be executed in a computer system such as a set of computer-executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in a different order than here.

[0047] The present invention discloses a hierarchical collaborative confluence control method for mixed traffic flow, belonging to the field of autonomous driving technology, including: establishing a collaborative control strategy framework based on a hierarchical structure, decomposing the problem into a confluence sorting layer and a motion control layer to improve the optimal control solution speed; constructing a 0-1 integer programming model, comprehensively considering traffic efficiency, safety and driving comfort, and transforming the confluence sorting problem into a shortest path search problem; designing a consistency controller considering communication delay, which can use the state information of all preceding vehicles for control decision-making, calculate the optimal control output and plan the future longitudinal movement of the autonomous vehicle; analyzing the local and platoon linear stability condition criteria and establishing parameter setting criteria. The present invention can complete collaborative confluence by controlling autonomous vehicles in mixed traffic flow, improving the overall traffic efficiency and safety in the confluence area.

[0048] The hierarchical collaborative confluence control method for mixed traffic flow provided by the present invention includes the following steps:

[0049] S1. Construct an optimized confluence order model and construct a physical network of the platoon; establish a cost function to determine the cost between all two sides in the optimization model; establish a 0-1 integer programming model to solve the optimized confluence order of all vehicles in the confluence area.

[0050] S2. Design a vehicle motion planning controller and construct a communication network of the platoon; establish a vehicle spacing strategy and a dynamic model; design a consistency controller considering communication delay and calculate the optimal longitudinal acceleration control quantity.

[0051] S3. Analyze the stability of the mixed - flow vehicle platoon cooperative control system, including local stability and platoon linear stability; use stability criteria for analysis, provide stability conditions; provide gain coefficient setting criteria, and the controller features high efficiency, stability, and robustness.

[0052] S4. Collect the real - time state information of all vehicles within the highway merging area as the input information for trajectory planning; calculate the planned trajectory of the vehicle at the next moment based on the real - time state information of the vehicle and update the vehicle state.

[0053] Preferably, the said S1 includes:

[0054] S11. For each edge under the merging order optimization model the cost function is:

[0055]

[0056] In the formula, the cost function consists of three parts. represents the traffic efficiency index. represents the driving comfort index. represents the traffic safety index. represents the weight factor reflecting human driving style.

[0057] Predict the motion state (position speed and acceleration ) of the vehicle to estimate . The motion trajectory of the autonomous vehicle is calculated by the controller proposed in the present invention, and the motion trajectory of the human - driven vehicle is calculated using other car - following models.

[0058] S12. The merging order optimization integer programming model is:

[0059]

[0060]

[0061] In the formula, represents the objective function, which represents the total cost of the given merging order. represents the decision variable, which represents whether to select edge , if edge is selected, then , otherwise .

[0062] The model needs to satisfy real-world constraints: Vehicles in the same lane cannot overtake, that is, the passing order of the vehicle in front in the same lane takes precedence over the vehicle behind; the sum of the selected edges must be equal to the total number of vehicles minus one, that is, the numbers of all vehicles in the control area must be in the final merging sequence; vehicles except the first (or last) vehicle on each lane must have an out-edge (or in-edge); each vehicle can have at most one out-edge or in-edge; there should be no sub-circuits in the sequence; only one of the two intersecting edges can be selected.

[0063] Preferably, the S2 includes:

[0064] S21. Construct a communication network for the platoon in the merging area: The platoon consists of vehicles, including a platoon leader vehicle (numbered 0), and following vehicles. The platoon leader vehicle 0 can move freely and cannot be controlled. The communication topology of the following vehicles is modeled as a directed graph .

[0065] represents the set of nodes controlled by the proposed merging algorithm. is the set of edges. is a weighted adjacency matrix used to represent whether there is communication between two vehicles, where is the edge 's weight. , if and only if (that is, vehicle can receive information from vehicle ); otherwise, . The Laplacian matrix is calculated by , where , is a diagonal matrix composed of all . represents the in-degree of node .

[0066] S22. Establish a vehicle spacing strategy and a vehicle dynamics model: To ensure the overall stability and safety of the traffic flow, the vehicle spacing needs to be dynamically adjusted according to factors such as the vehicle's current speed, reaction time, and road conditions. The vehicle dynamics model models the non-linear vehicle dynamics in the real world.

[0067] Adopt a constant headway as the vehicle spacing strategy to reduce the generation and propagation of traffic waves.

[0068]

[0069]

[0070] In the formula, Indicates the desired constant headway. Represents the vehicle At the moment Longitudinal speed of, Represents the vehicle Length of, Represents the minimum distance between vehicles, Represents the vehicle Desired distance from the leading vehicle 0 of the platoon, Represents the vehicle Desired headway from the leading vehicle of the platoon.

[0071] The vehicle dynamics model models the non-linear vehicle dynamics in the real world.

[0072] The vehicle dynamics model is:

[0073]

[0074] Wherein, Represents the jerk, Represents the desired acceleration generated by the consensus controller, Represents the vehicle Time delay to achieve the acceleration, Represents the ratio of the desired acceleration that can be achieved.

[0075] S23. The goal of the consensus controller is that as time approaches infinity, for the vehicle , to make the error , And Gradually approach zero:

[0076]

[0077]

[0078]

[0079] The state space system equation is:

[0080]

[0081] Wherein, Represents the system matrix, Represents the input matrix, Represents the external disturbance matrix. Represents the acceleration of the leading vehicle of the platoon. The S3 includes:

[0082] S31. Local stability analysis: Set appropriate gain coefficients in the controller. When facing external disturbances or internal control errors, adjust its own state to restore to the desired trajectory or speed, ensuring that the system remains stable under the proposed control strategy.

[0083] S32. Queue linear stability analysis: Set appropriate gain coefficients in the controller to ensure that the acceleration or deceleration of the leading vehicle does not cause greater disturbances in the following vehicles, but gradually decreases as the vehicle is transmitted, thereby reducing the oscillation and congestion of the vehicle fleet and improving the stability and safety of the vehicle fleet.

[0084] Preferably, the S4 includes:

[0085] S41. Real-time data acquisition: Collect the real-time state information of all vehicles within the highway merge area, including the current longitudinal position, longitudinal speed, and longitudinal acceleration, as the dynamic input information for cooperative merge control; static input information such as vehicle type and lane to which the vehicle belongs; and road structure information such as merge area length, road width, and road boundaries.

[0086] S42. Vehicle state update: Control the vehicle according to the calculated optimal longitudinal acceleration and update the longitudinal position and longitudinal speed of all vehicles at the next moment.

[0087] The above technical solutions of the present invention will be described in detail in conjunction with the relevant drawings:

[0088] As Figure 1 shown, in this embodiment, a hierarchical cooperative merge control method for mixed traffic flow is provided, including:

[0089] S1. Construct a merge sequence optimization model and construct a physical network of the vehicle fleet; establish a cost function to determine the cost between all two sides in the optimization model; establish a 0-1 integer programming model to solve the merge sequence optimization of all vehicles in the merge area.

[0090] S2. Design a vehicle motion planning controller and construct a communication network of the vehicle fleet; establish a vehicle spacing strategy and a dynamic model; design a consensus controller considering communication delay and calculate the optimal longitudinal acceleration control amount.

[0091] S3. Analyze the stability of the cooperative control system of the mixed-flow vehicle fleet, including local stability and fleet linear stability; use stability criteria for analysis to provide stability conditions; provide gain coefficient setting criteria, and the controller has the characteristics of high efficiency, stability, and high robustness.

[0092] S4. Collect the real-time state information of all vehicles within the highway merge area as the input information for trajectory planning; calculate the planned trajectory of the vehicle at the next moment based on the real-time state information of the vehicle and update the vehicle state.

[0093] Specifically, S1 includes:

[0094] S11. For each edge under the confluence order optimization model the cost function is:

[0095]

[0096] In the formula, the cost function consists of three parts, represents the traffic efficiency index, represents the driving comfort index, represents the traffic safety index. represents the weight factor reflecting the human driving style.

[0097]

[0098]

[0099]

[0100] In the formula, , . , and represent the longitudinal position, speed and acceleration of vehicle . In, if the longitudinal speed is closer to the desired speed , the traffic efficiency cost becomes smaller. becomes smaller with the decrease of the longitudinal acceleration. represents the safe vehicle spacing. represents the safety cost. If the longitudinal vehicle spacing is greater than the safe distance, it means that there is no collision risk between vehicles at this time, which can be represented by . If the longitudinal vehicle spacing is less than the safe distance, there is a collision risk at this time, and the risk will increase with the increase of , which can be represented as .

[0101] Predict the motion state of the vehicle (position , speed and acceleration ) to estimate . The motion trajectory of the autonomous vehicle is calculated by the controller proposed in the present invention, and the motion trajectory of the human-driven vehicle is calculated by using other car-following models.

[0102] The motion trajectory prediction model of the human-driven vehicle can adopt the 2D-IDM model:

[0103]

[0104]

[0105]

[0106]

[0107] In the formula, represents the maximum acceleration, represents the safe deceleration. represents the expected spacing of the vehicle and represents the minimum spacing. represents the expected time interval, represents the target time interval. and represent two random numbers between 0 and 1. represents the maximum change rate of the expected time interval, and represent the minimum and maximum time intervals. represents the random probability.

[0108] The integer programming model for optimizing the merging sequence S12 is:

[0109]

[0110]

[0111] In the formula, represents the objective function, which is the total cost of a given merging sequence. represents the decision variable, indicating whether to select edge , if edge is selected, then , otherwise .

[0112] The model needs to satisfy real-world constraints: Vehicles in the same lane cannot overtake, that is, the passing order of the vehicle in front in the same lane takes precedence over the vehicle behind; the sum of the selected edges must be equal to the total number of vehicles minus one, that is, the numbers of all vehicles in the control area must be in the final merging sequence; each vehicle except the first (or last) vehicle on each lane must have an out-edge (or in-edge); each vehicle can have at most one out-edge or in-edge; there should be no sub-circuits in the sequence; only one of the two intersecting edges can be selected.

[0113]

[0114]

[0115]

[0116]

[0117]

[0118]

[0119]

[0120]

[0121]

[0122]

[0123]

[0124]

[0125]

[0126] wherein, and represent lanes and are the numbers of the last vehicles on the lanes. and represent auxiliary variables and represent the order of nodes in the sequence. , , , represent the criteria for whether the two sides intersect. Among them, , represents the set of lanes, represents the set of vehicles. , , , represent four different vehicle numbers, , , , represent four different lane numbers. The feasible and infeasible merge sorting situations are as shown in Figure 2 shown.

[0127] Specifically, the S2 includes:

[0128] S21. Construct a communication network for the platoon in the merge area: The platoon consists of vehicles, including a platoon leader vehicle (numbered 0), and following vehicles. The platoon leader vehicle 0 can move freely and cannot be controlled. The communication topology of the following vehicles is modeled as a directed graph .

[0129] Represents the set of nodes that can be controlled by the proposed merging algorithm. Is the set of edges. Is a weighted adjacency matrix used to represent whether there is communication between two vehicles, where Is the edge The weight of. , if and only if (i.e., vehicle Can receive information from vehicle ); otherwise, . The Laplacian matrix Can be obtained by Calculated, where , is a diagonal matrix composed of all Composed. Represents the in-degree of node .

[0130] Specifically, the following gives an example of a communication topology network described by the Laplacian matrix: Suppose there is a convoy of 4 vehicles, and its communication topology is as follows: Vehicle 1 can receive information from Vehicle 2 and Vehicle 3. Vehicle 2 can receive information from Vehicle 3. Vehicle 3 can receive information from Vehicle 4. Vehicle 4 does not receive information from any vehicle.

[0131] According to the above communication topology, the weighted adjacency matrix And the in-degree matrix Can be constructed.

[0132] The weighted adjacency matrix Is: , the in-degree matrix Is: . Therefore, the Laplacian matrix Under this communication topology is: .

[0133] S22. Establish a vehicle spacing strategy and a vehicle dynamics model: To ensure the overall stability and safety of the traffic flow, the vehicle spacing needs to be dynamically adjusted according to factors such as the vehicle's current speed, reaction time, and road conditions.

[0134] Adopt a constant headway as the vehicle spacing strategy to reduce the generation and propagation of traffic waves.

[0135]

[0136]

[0137] In the formula, Denotes the desired constant headway. Denotes the vehicle At time The longitudinal speed of Denotes the vehicle The length of Denotes the minimum distance between vehicles Denotes the vehicle The desired distance from the leading vehicle of the platoon Denotes the vehicle The desired headway from the leading vehicle of the platoon, with subscript 0 denoting the leading vehicle of the platoon.

[0138] The vehicle dynamics model models the non - linear vehicle dynamics in the real world:

[0139]

[0140] Wherein, Denotes the jerk Denotes the desired acceleration generated by the consensus controller Denotes the vehicle The time delay for achieving the acceleration Denotes the ratio of the achievable desired acceleration.

[0141] S23. The goal of the consensus controller is that as time approaches infinity, for the vehicle , to make the errors , And Gradually approach zero. The longitudinal position error Between the vehicle and the desired position And the longitudinal speed error Between the autonomous vehicle and the leading vehicle of the platoon:

[0142]

[0143]

[0144] Wherein, , Denotes the speed of the leading vehicle.

[0145] Specifically, the S3 includes:

[0146] S31. Local stability analysis: Set appropriate gain coefficients in the controller. In the face of external disturbances or internal control errors, adjust its own state to restore to the desired trajectory or speed, ensuring that the system remains stable under the proposed control strategy.

[0147] The Lyapunov-Razumikhin theorem can be used for analysis. When the following conditions are satisfied, the mixed-flow vehicle platoon has global asymptotic stability:

[0148]

[0149] Among them, represents the row and column element indices of the matrix, represents the total number of states, is the coupling gain, represents a constant and , represents the eigenvalue of the Laplacian matrix, represents the system matrix, represents the input matrix, represents the gain vector.

[0150] S32. Analysis of queue linear stability: Set the gain coefficients in the appropriate controller to ensure that the acceleration or deceleration of the leading vehicle does not cause greater disturbances in the following vehicles, but gradually decreases as the vehicle transmission progresses, thereby reducing the oscillation and congestion phenomena of the platoon and improving the stability and safety of the platoon.

[0151] The final queue linear stability theorem can be used for analysis. When the following conditions are satisfied, the mixed-flow vehicle platoon has queue linear stability:

[0152]

[0153]

[0154]

[0155]

[0156] Among them, , , respectively represent the position error, speed error, and acceleration error gain coefficients. represents the communication delay, represents the frequency.

[0157] The feasible region of the gain coefficients is as Figure 3 shown.

[0158] Specifically, the said S4 includes:

[0159] S41. Real-time data collection: Collect the real-time status information of all vehicles within the highway merging area, including the current longitudinal position, longitudinal speed, and longitudinal acceleration, as the dynamic input information for cooperative merging control; static input information such as vehicle type and the lane to which the vehicle belongs; and road structure information such as the length of the merging area, road width, and road boundaries.

[0160] S42. Vehicle status update: Control the vehicle according to the calculated optimal longitudinal acceleration and update the longitudinal position and longitudinal speed of all vehicles at the next moment.

[0161] The trajectory planning is a dynamic process. Generate the trajectories of all vehicles at time t + 1 based on the status at the current time t. The generated trajectories are as Figure 4 shown, and each trajectory curve corresponds to a vehicle.

[0162] Beneficial effects of this embodiment:

[0163] In this embodiment, a cost function considering traffic efficiency, comfort, and safety is established to calculate the merging sequence cost between any two vehicles. A 0-1 integer programming model is introduced to optimize the passing sequence of vehicles in the merging area. The established model can improve regional traffic efficiency and safety, and the algorithm calculation cost is relatively low. A consensus controller with communication delay is designed, and the state error between the vehicle and all the preceding vehicles is used as the control input. The proposed controller can stabilize the mixed traffic flow and improve the robustness. The setting criteria for the gain coefficient of the controller are analyzed to achieve efficient and stable control, and at the same time effectively cope with the influence of speed fluctuations of human-driven vehicles in the mixed vehicle fleet.

[0164] The above is only a preferred specific embodiment of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by those skilled in the art within the technical scope disclosed in the present application should be covered by the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. Hierarchical collaborative merging control method for mixed traffic flow, characterized in that, Including: Construct a utility function according to the interaction characteristics of vehicles in mixed traffic flow, and construct an integer programming model for optimizing the merging order according to the utility function; Construct a platoon communication network according to the real-time state information of vehicles, construct an inter-vehicle distance strategy and a vehicle dynamics model according to the platoon communication network, and construct a consensus controller according to the vehicle dynamics model; wherein the gain coefficient in the consensus controller is constrained by local stability and queue linear constraint stability analysis; Obtain the real-time information of the vehicles, optimize the real-time information of the vehicles through the integer programming model for optimizing the merging order to obtain the optimal merging order, and calculate the vehicle control information according to the optimal merging order through the consensus controller according to the real-time information of the vehicles.

2. The method according to claim 1, wherein The utility function is: Among them, represents the traffic efficiency index, represents the driving comfort index, represents the traffic safety index, , , represent the weight factors, represents the total cost.

3. The method according to claim 2, wherein The obtaining process of different indexes in the utility function includes: Among them, , and represent the longitudinal position, speed and acceleration of the vehicle . represents the desired speed, represents the safe vehicle distance, represents the time interval, represents the initial time, , represents the vehicle label, represents all vehicles.

4. The method according to claim 3, wherein The integer programming model for optimizing the merging order is: Among them, represents the total cost of a given merging order, represents a decision variable indicating whether to select edge , if edge is selected, then , otherwise .

5. The method according to claim 1, wherein The construction process of the platoon communication network includes: Construct a platoon communication network according to the real-time state information of the vehicles, wherein the platoon communication network is a directed graph structure, including a node set, an edge set and a weighted adjacency matrix, wherein the node set includes the control vehicles in the platoon, the edge set includes the relationships between vehicles, and the weighted adjacency matrix represents whether there is communication between vehicles; Describe the platoon communication network through the Laplacian matrix.

6. The method according to claim 1, wherein The inter-vehicle distance strategy is: Among them, represents the desired constant headway, represents the vehicle at time longitudinal speed, represents the vehicle length, represents the minimum spacing between vehicles, represents the vehicle label, represents the node set, represents the vehicle desired spacing from the leading vehicle of the platoon, represents the vehicle desired headway from the leading vehicle of the platoon.

7. The method according to claim 1, wherein The vehicle dynamics model is: Among them, represents jerk, represents acceleration, represents the desired acceleration generated by the consistency controller, represents the vehicle realizes the time delay of acceleration, represents the ratio of the achievable desired acceleration, represents the vehicle number.

8. The method according to claim 5, wherein The consensus controller includes: Control the longitudinal position, longitudinal speed and acceleration of the control vehicle, wherein the error between the longitudinal position and the desired position, the error between the longitudinal speed and the desired speed, and the acceleration gradually approach zero.

9. The method according to claim 1, wherein The constraint process of the gain coefficient in the consensus controller includes: Conduct local stability analysis by adopting the Lyapunov-Razumikhin theorem to obtain local gain coefficient constraints; Conduct queue linear stability analysis through the final queue linear stability theorem to obtain queue linear gain coefficient constraints; Constrain the gain coefficient by obtaining local gain coefficient constraints and queue linear gain coefficient constraints.

Images