Signal intersection intelligent vehicle cooperative traffic model prediction control method based on vehicle-road cloud interaction

Abstract

The invention relates to a signalized intersection intelligent vehicle cooperative traffic model prediction control method based on vehicle and road cloud interaction, and belongs to the technical field of intelligent traffic, and the method comprises the following steps: S1, constructing a multi-lane signalized intersection traffic scene model, including defining lane positions and serial numbers of each lane vehicle and adjacent vehicles; s2, establishing a longitudinal tracking state model and a transverse motion model of a vehicle in the scene, and forming a vehicle longitudinal and transverse coupling state model; s3, designing an optimization problem under a model predictive control (MPC) framework, and constructing a multi-objective cost function which comprehensively considers vehicle driving efficiency, economy, traffic feasibility and speed excitation; s4, setting a vehicle speed constraint, an acceleration constraint, a safe following distance constraint and a lane change collision safety constraint in the MPC optimization problem; and S5, solving an MPC optimization problem based on rolling horizon optimization, and outputting a speed control sequence and a lane changing decision of the main vehicle in the prediction horizon as a control strategy of the vehicle.

Description

Technical Field

[0001] The present invention belongs to the field of intelligent transportation technology and relates to a predictive control method for an intelligent automobile cooperative traffic model at a signal intersection based on vehicle-road-cloud interaction. Background Art

[0002] With the rapid growth of urbanization and the number of motor vehicles, urban traffic congestion, energy consumption, and emissions are becoming increasingly serious. As key nodes in the transportation network, intersections have a decisive impact on the overall performance and energy consumption of the transportation network. Traditional signal control methods are usually based on macroscopic traffic flow parameters and fail to consider the microscopic motion characteristics of vehicles at intersections and the interaction between multiple lanes. Intelligent connected vehicles (CAVs) provide new ideas for refined control of intersection traffic through information exchange between vehicles and between vehicles and roads. Existing studies have used signal phase information (SPaT) to plan vehicle speeds, but most focus on single-lane speed optimization and lack consideration of vehicle lane-changing behavior in multi-lane environments. In addition, existing methods often separate longitudinal control and lateral decision-making, making it difficult to achieve joint optimization of the two, resulting in limited improvements in vehicle efficiency and fuel economy at intersections. Therefore, a new control strategy is needed to simultaneously optimize vehicle speed and lane-changing decisions at signalized intersections. Summary of the Invention

[0003] In light of this, the present invention aims to provide a predictive control method for intelligent vehicle cooperative traffic at signalized intersections based on vehicle-road-cloud interaction. This method aims to achieve coupled longitudinal and lateral control of individual vehicles in multi-lane intersection scenarios by integrating inter-vehicle information exchange, signal timing information, and multi-objective optimization. This method improves intersection efficiency and energy utilization while ensuring safety, addressing the inability of existing technologies to simultaneously optimize vehicle speed and lane change decisions.

[0004] In order to achieve the above object, the present invention provides the following technical solutions:

[0005] A predictive control method for intelligent vehicle cooperative traffic model at a signalized intersection based on vehicle-road-cloud interaction includes the following steps:

[0006] S1: Construct a traffic scenario model for a multi-lane signalized intersection, including defining the lane positions and sequence numbers of vehicles in each lane and adjacent vehicles;

[0007] S2: Establish the longitudinal tracking state model and lateral motion model of the vehicle in the scene, and form the longitudinal and lateral coupling state model of the vehicle;

[0008] S3: Design an optimization problem within the MPC framework and construct a multi-objective cost function that comprehensively considers vehicle efficiency, economy, traffic feasibility, and speed incentives.

[0009] S4: Set vehicle speed constraints, acceleration constraints, safe following distance constraints, and lane change collision safety constraints in the MPC optimization problem;

[0010] S5: Solve the MPC optimization problem based on the receding horizon optimization, and output the speed control sequence and lane change decision of the main vehicle in the prediction horizon as the vehicle control strategy.

[0011] Furthermore, in step S1, the position of the vehicle on the road is defined as (i, j), where i is the lane index of the vehicle and j is the serial number of the vehicle in the lane; the main vehicle is recorded as (m, 0), the vehicle in front of the main vehicle is recorded as (m, 1), the following vehicles closest to the main vehicle in the left and right lanes of the main vehicle are recorded as (l, 0) and (r, 0), respectively, and the front vehicles in the left and right lanes are recorded as (l, 1) and (r, 1), respectively; the possible lane change index is described by δ(k) = {-1, 0, 1}, where -1 indicates a left lane change, +1 indicates a right lane change, and 0 indicates no lane change; vehicles in the leftmost lane cannot change lanes to the left, and vehicles in the rightmost lane cannot change lanes to the right; the lane index i of each vehicle is recorded, where N represents the total number of lanes. Right-turning vehicles do not participate in the control strategy, and only left-turning and straight-moving vehicles are considered.

[0012] Furthermore, in step S2, a discrete-time dynamic model is established for the longitudinal tracking state of the vehicle. The speed of the jth CAV in the i-th lane at time t+1 is determined by its speed and acceleration at time t. The position of the jth CAV in the i-th lane at time t+1 is related to its acceleration and speed at time t. Let S, V, and a be the position, speed, and acceleration of the CAV, respectively. The relationship between them is expressed as:

[0013] V (i,j),t+1 =V (i,j),t +a (i,j),t ×Δt

[0014]

[0015] Where V (i,j),t+1 is the speed of the j-th CAV in the i-th lane at time t+1, V (i,j),t is the speed of the j-th CAV in lane i at time t, a (i,j),t The acceleration of the jth CAV in the i-th lane at time t, Δt is the sampling time; S (i,j),t+1 is the position of the jth CAV in the i-th lane at time t+1, S (i,j),t is the position of the j-th CAV in the i-th lane at time t;

[0016] The longitudinal state space formula of a single CAV is:

[0017] x i,j (k+1)=A·xi,j (k)+B·u i,j (k)

[0018] x i,j (k)=[S i,j (k),V i,j (k)] Τ

[0019] u i,j (k)=[a i,j (k)]

[0020]

[0021] where x i,j is the state vector of vehicle (i, j), u i,j is the vehicle (i, j) control vector, k is the discrete time index, and t is the time step.

[0022] Furthermore, in step S2, a discrete-time dynamic model is established for the lateral motion of the vehicle, and the state variables and control inputs are:

[0023]

[0024] Among them, y i,j is the lateral displacement of vehicle i relative to the lane centerline, with left deviation being positive and right deviation being negative; is the lateral velocity, is the lateral acceleration;

[0025] The sampling time is Δt, and the discretized model is:

[0026]

[0027] Furthermore, in step S2, the longitudinal and lateral models of the vehicle are coupled to construct an extended state space model of the vehicle. The overall state space is expanded to:

[0028]

[0029] in represents the state vector of the longitudinal motion of vehicle (i, j), represents the longitudinal velocity of vehicle (i, j);

[0030] The coupled dynamic equations are expressed as follows:

[0031]

[0032] Among them A lon represents the longitudinal motion state matrix, A lat represents the lateral motion state matrix, B lonrepresents the longitudinal motion input matrix, B lat represents the lateral motion input matrix, Represents the total lateral and longitudinal control input of vehicle (i, j).

[0033] Furthermore, the multi-objective cost function in step S3 includes a lane tracking term, a speed error term, a traffic feasibility term, and an acceleration term;

[0034] The lane tracking item is used to make the host vehicle as close to the predetermined target lane as possible;

[0035] The speed error term represents the deviation between the vehicle’s current speed and the desired speed;

[0036] The traffic feasibility item introduces the average speed information of different lanes to encourage the main vehicle to change lanes to a lane with higher traffic efficiency;

[0037] The acceleration term is used to measure the energy consumption caused by acceleration;

[0038] Each part is multiplied by the corresponding weight coefficient and combined to form the overall objective function. Taking into account the performance and control objectives, the objective function of the optimization problem is expressed as:

[0039]

[0040] Where t is the current time from the start of the simulation time, k is the kth step beyond the prediction horizon, and w l ,w v ,w t ,w a is the weight factor, M represents the number of prediction steps, y(k) represents the lane where the vehicle is located at the kth step, and y ref,p (k) represents the target lane expected at step k, v m,0 (k) represents the current speed of the host vehicle at step k, represents the target speed expected by the main vehicle in lane p∈δ(k) at step k, represents the target speed expected by the main vehicle in lane p∈δ(k) at step k, It represents the maximum average target speed of different lanes at time t within the predicted horizon, represents the acceleration of the main vehicle, p represents the lane-changing behavior, and δ(k) represents the lane-changing parameter, which describes the vehicle's movement intention.

[0041] Furthermore, the vehicle speed constraint, acceleration constraint, safe following distance constraint, and lane change collision safety constraint in step S4 are specifically:

[0042] The vehicle's speed must not exceed the set maximum speed and must not be lower than the set minimum speed; the vehicle's acceleration must not exceed the set maximum value and must not be lower than the set minimum value; the main vehicle and the vehicle in front must maintain a headway of at least the set minimum time; during the lane change process, a safe distance must be maintained between the main vehicle and vehicles in the left and right lanes to avoid collisions. The constraints are expressed as follows:

[0043] 0≤v m,0 ≤v max

[0044] a min ≤a m,0 ≤a max

[0045] Among them, v m,0 is the vehicle speed, v max is the maximum speed limit of the lane, a min The minimum acceleration of the main vehicle, a max is the maximum acceleration of the main vehicle;

[0046]

[0047] Among them, x m,1 is the position of the preceding vehicle, x m,0 For the main vehicle position, is the minimum headway, R0 is the minimum gap between the main vehicle and the preceding vehicle;

[0048] (x i,j (k)-x m,0 (k)) 2 ≥δ i,j (k)·R c

[0049]

[0050] where x i,j (k) is the state variable of vehicle (i, j) at step k, x m,0 (k) is the state variable of the main vehicle at step k, δ i,j (k) is the lane-changing discrete action, R c is the minimum gap between the main vehicle and the preceding vehicle in different lanes, L i,j (k) represents the index of the lane where the k-th vehicle (i, j) is located, L m,0 Indicates the lane index of the host vehicle.

[0051] Furthermore, in step S5, a rolling time domain optimization solution method is used to predict the motion state of the main vehicle in the future time domain by collecting traffic signal phase and timing SPaT information and real-time status information of surrounding vehicles; the quadratic programming QP numerical optimization algorithm is used to minimize the cost function, and the solution is iteratively solved in the rolling time domain to obtain the optimal acceleration sequence and lane change decision of the main vehicle; only the first control input of the optimization sequence is applied to the vehicle, and the optimization process is updated at the next moment based on the newly acquired status information; the green wave speed V of the vehicle in the green light state at time t is signal (t) is represented by the following formula:

[0052]

[0053] Where d(t) represents the distance between the current vehicle and the signal intersection, t rem is the remaining time of the current phase, represents the maximum speed of green wave at time t, V max Indicates the maximum allowed speed, T g and T r They are the time when the current lane opens green light and the time when the current lane opens red light in each signal cycle, k = 0, 1, 2, 3, ..., ∞ represents the acceptable green light time sequence, which will be satisfied in the specific calculation The first window of is used as the basis for the final k value;

[0054] The target speed of the main vehicle at time t under p (p∈δ(t)) is determined as a function of the headway:

[0055]

[0056] Among them, α1 and α2 are constants for adjusting the speed amplitude, (c, 1) means the front vehicle is driving in different lanes, l is the left lane, m is the main vehicle lane, r is the right lane, s c,1 is the position of the preceding vehicle, v c,1 is the speed of the preceding vehicle, is the reference time, i.e. the ideal headway; s m,0 (t) is the position of the main vehicle at time t, v m,0 (t) is the main vehicle speed at time t; the headway calculation formula is as follows:

[0057]

[0058] where s c,0 (t) is the position of the main vehicle in different lanes at time t, v c,0 (t) is the speed of the main vehicle in different lanes at time t;

[0059] At time k, the target velocity on the M-step prediction horizon is expressed as:

[0060]

[0061] In the forecast horizon, the target expected speed Changes over time, It is the average value of the target expected speed under the lane change behavior p within the prediction horizon:

[0062]

[0063] is the maximum value of the target speed between different lanes:

[0064]

[0065] in, is the average speed in different lanes;

[0066] Target acceleration a t,p (k) The target speed between two steps in each prediction interval is expressed as:

[0067]

[0068] By using the above parameters, the target speed on the predicted horizon is determined, and the cost function is optimized to obtain the minimum cost under different lane change operations on the predicted horizon, and then the optimal Calculate the optimal lane and get the optimal acceleration

[0069] The beneficial effects of the present invention are as follows: the present invention fully considers the multi-lane environment of the intersection and the interaction between vehicles, and realizes the coupled optimization of vehicle speed control and lane change decision. The method integrates signal phase information with the state of surrounding vehicles under the framework of model predictive control, takes into account both traffic efficiency and energy economy through multi-objective optimization, and ensures the feasibility of the control strategy through strict safety constraints. Simulation results show that the control strategy of the present invention can significantly improve the efficiency of vehicles passing through intersections, reduce parking waiting time, and reduce vehicle energy consumption. In addition, the method has good robustness and adaptability, and can provide an effective technical means for intelligent traffic control at signalized intersections.

[0070] Other advantages, objects, and features of the present invention will be described in part in the following description and, in part, will be apparent to those skilled in the art upon examination of the following description or may be learned from practice of the present invention. The objects and other advantages of the present invention may be realized and obtained through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention will be described in detail below with reference to the accompanying drawings, in which:

[0072] Figure 1 This is a flow chart of the traffic optimization model predictive control of the present invention;

[0073] Figure 2 The detailed MPC process used in this example;

[0074] Figure 3 Modeling the traffic scenario of a multi-lane road in this embodiment;

[0075] Figure 4 Schematic diagram of the basic idea of ​​speed solution to the optimization problem in this embodiment. DETAILED DESCRIPTION

[0076] The following describes the embodiments of the present invention by means of specific examples, and those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present invention, and the following embodiments and features in the embodiments can be combined with each other without conflict.

[0077] It should be noted that the illustrations provided in the following embodiments are merely schematic illustrations of the basic concept of the present invention. Therefore, the illustrations only show components related to the present invention and are not drawn according to the number, shape, and size of components in actual implementation. In actual implementation, the type, quantity, and proportion of each component may be changed arbitrarily, and the component layout may also be more complex.

[0078] In the following description, numerous details are discussed to provide a more thorough explanation of the embodiments of the present invention. However, it will be apparent to those skilled in the art that the embodiments of the present invention may be practiced without these specific details. In other embodiments, well-known structures and devices are shown in block diagram form rather than in detail to avoid obscuring the embodiments of the present invention.

[0079] Example 1:

[0080] like Figure 1-4 As shown, the present invention provides a predictive control method for intelligent vehicle cooperative traffic model at a signal intersection based on vehicle-road-cloud interaction, comprising the following steps:

[0081] In this embodiment, it is assumed that each entrance direction of a signalized intersection has multiple lanes, including left-turn lanes, straight lanes, and right-turn lanes. The position of a vehicle on the road is defined as (i, j), where i is the lane index of the vehicle (numbered from left to right) and j is the serial number of the vehicle in that lane. This invention focuses on vehicles going straight or turning left. Right-turning vehicles are not subject to signal control and do not participate in the control strategy. Assume that the main vehicle (the target vehicle to be optimized) is in lane numbered i = m, which is labeled (m, 0), and the preceding vehicle is labeled (m, 1). To achieve collision prevention during lane changes, the nearest following vehicles in the lanes immediately to the left and right of the main vehicle are taken into consideration and are denoted as (l, 0) and (r, 0), where l = m-1 (left lane) and r = m+1 (right lane). Similarly, the leading vehicles in the lanes to the left and right of the main vehicle are denoted as (l, 1) and (r, 1), respectively. To characterize lane change decisions, -1 indicates a left lane change, +1 indicates a right lane change, and 0 indicates no lane change. Vehicles in the leftmost lane cannot change lanes to the left, and vehicles in the rightmost lane cannot change lanes to the right. Through the above settings, the lane positions and serial numbers of the main vehicle and surrounding key vehicles can be fully described, providing a basis for subsequent control strategy design.

[0082] Based on the above scenario modeling, a vehicle kinematic model is established. In this embodiment, it is assumed that all vehicles are autonomous vehicles that can obtain surrounding environment information in real time with negligible response delay. First, a longitudinal tracking model of the vehicle is established. The speed of the jth CAV in the i-th lane at time t+1 is determined by its speed and acceleration at time t. The position of the jth CAV in the i-th lane at time t+1 is related to its acceleration and speed at time t. Let S, V, and a define the position, speed, and acceleration of the CAV respectively. The relationship between them can be expressed as:

[0083] V (i,j),t+1 =V (i,j),t +a (i,j),t ×Δt

[0084]

[0085] Where Δt is the sampling time. Therefore, the longitudinal state space formula of a single CAV is:

[0086] x i,j (k+1)=A·x i,j (k)+B·u i,j (k)

[0087] x i,j (k)=[S i,j (k),V i,j (k)] Τ

[0088] u i,j (k)=[ai,j (k)]

[0089]

[0090] where x i,j is the state vector of vehicle (i, j), u i,j is the vehicle (i, j) control vector, k is the discrete time index, and t is the time step.

[0091] The lateral displacement and lateral velocity form a state vector, which can be written as a similar linear discrete model. The state variables and control inputs are:

[0092]

[0093] Among them, y i,j The lateral displacement of vehicle i relative to the lane centerline, with left deviation being positive and right deviation being negative; lateral speed, Lateral acceleration.

[0094] The sampling time is Δt, and the discretized model is:

[0095]

[0096] By combining the longitudinal and lateral models, we can obtain a coupled state space model of the vehicle, which simultaneously describes the longitudinal following and lateral lane-changing behaviors of the vehicle in an intersection environment. The overall state space can be expanded to:

[0097]

[0098] The coupled dynamic equations can be expressed as follows:

[0099]

[0100] Based on the kinematic model, a trajectory optimization method based on model predictive control (MPC) is designed. MPC can predict the vehicle motion within a certain time domain in the future at each control moment, and solve the constrained optimization problem online to generate the optimal control sequence.

[0101] The present invention defines a multi-objective cost function under the MPC framework, comprehensively considering multiple performance indicators of vehicles passing through intersections, including: driving efficiency, energy economy, safety and speed incentives. Specifically, the cost function includes the following components: the first part is the lane tracking term, which is used to keep the main vehicle in the predetermined target lane as much as possible; the second part is the speed error term, which is used to measure the difference between the current speed of the vehicle and the expected speed. The expected speed can be determined in real time based on traffic signals and the status of the vehicle in front; the third part is the traffic feasibility term, which encourages vehicles to change lanes to lanes with higher traffic efficiency by introducing the average driving speed information of different lanes; the fourth part is the acceleration term, which is used to reflect the energy consumption required for acceleration or deceleration. Each part is multiplied by the corresponding weight and superimposed to form the final optimization objective function J. The objective function of the optimization problem can be expressed as:

[0102]

[0103] Where t is the current time from the start of the simulation time, k is the kth step beyond the prediction horizon, and w l ,w v ,w t ,w a is the weight factor.

[0104] During objective function optimization, vehicle driving constraints must be met, including: vehicle speed must not exceed the road speed limit and must not fall below a preset minimum speed; acceleration must not exceed preset maximum and minimum values; the lead vehicle must maintain a headway of at least a given minimum between it and the vehicle ahead; and during lane changes, the lead vehicle must maintain a sufficient safety gap with vehicles in adjacent lanes to avoid collisions. These constraints ensure that the optimization results are physically feasible and meet safety requirements. The corresponding constraints are as follows:

[0105] 0≤v m,0 ≤v max

[0106] a min ≤a m,0 ≤a max

[0107] Among them, v max is the maximum speed limit of the lane, a min Minimum acceleration of the host vehicle, a max The maximum acceleration of the host vehicle.

[0108]

[0109] in, is the minimum headway, and R0 is the minimum gap between the main vehicle and the preceding vehicle.

[0110] (x i,j (k)-xm,0 (k)) 2 ≥δ i,j (k)·R c

[0111]

[0112] In order to solve the above MPC optimization problem, this embodiment adopts a rolling time domain optimization strategy. At each control moment, by obtaining the signal light phase and timing information (SPaT) and the real-time status information of the surrounding vehicles, the movement of the main vehicle in the future time domain under different lane conditions is predicted. Specifically, the target speed sequence of the vehicle in different lanes is calculated based on the optimization results of the previous moment, and then the cost evaluation is performed for the three situations of the main vehicle staying in the current lane, driving to the left lane or the right lane. Numerical optimization algorithms such as quadratic programming (QP) are used to minimize the objective function J and meet the above constraints to obtain the optimal control sequence, including the acceleration control input and lane change action decision of the main vehicle. After the first control quantity in the optimization sequence is applied to the vehicle, the optimization is repeated at the next moment based on the newly acquired state information, thereby forming a closed-loop control and adjusting the vehicle trajectory in real time. The green wave speed V of the vehicle in the green light state at time t signal (t) can be expressed by the following formula:

[0113]

[0114] Among them, t rem is the remaining time of the current phase; t g and t r They are the time when the current lane opens green light and the time when the current lane opens red light in each signal cycle, k = 0, 1, 2, 3, ..., ∞ represents the acceptable green light time sequence, which will be satisfied in the specific calculation The first window of is used as the basis for the final k value.

[0115] The target speed of the main vehicle at time t under p (p∈δ(t)) is determined as a function of the headway:

[0116]

[0117] Among them, α1 and α2 are constants for adjusting the speed amplitude, (c,1) is the front vehicle driving in different lanes, s c,1 is the position of the preceding vehicle, v c,1 Speed ​​of the preceding vehicle, is the reference time (ideal headway), and the headway calculation formula is as follows:

[0118]

[0119] Therefore, at time k, the target velocity on the M-step prediction horizon can be obtained as follows:

[0120]

[0121] In the forecast horizon, the target expected speed Changes over time, It is the average value of the target expected speed under the lane change behavior p(p∈δ(t)) within the prediction horizon:

[0122]

[0123] Due to different lane change decisions, the speeds in different lanes are different. is the maximum value of the target speed between different lanes:

[0124]

[0125] in, is the average speed in different lanes.

[0126] Target acceleration a t,p (k) The target speed between two steps in each prediction interval is expressed as:

[0127]

[0128] With these parameters, the expected target speed is determined by the formula to determine the target speed on the prediction horizon. The cost function is optimized to obtain the minimum cost under different lane change operations on the prediction horizon. Then, in the optimal Calculate the optimal lane and get the optimal acceleration

[0129] Finally, the effectiveness of the method of the present invention was verified through simulation experiments. The SUMO microscopic traffic simulation platform was used to conduct tests in a typical two-way six-lane intersection scenario. By randomly generating traffic flows with different lane occupancy rates, a variety of vehicle density distribution scenarios were constructed, and the method of the present invention was compared with the traditional MPC method that only considers longitudinal speed optimization under different traffic flow scenarios. The results show that under various working conditions, the MPC control strategy based on vehicle interaction proposed in the present invention can significantly improve vehicle traffic efficiency, reduce vehicle traffic delays, and reduce fuel consumption. Especially under conditions of dense traffic flow, the method effectively alleviates congestion through intelligent lane change decisions, thereby improving the overall traffic efficiency of the intersection. These simulation results demonstrate the superiority of the technical solution of the present invention.

[0130] In summary, the present invention proposes a predictive control method for the collaborative traffic model of intelligent vehicles at signalized intersections that takes into account vehicle interaction and signal timing information. By constructing a vehicle dynamics model in a multi-lane environment and performing multi-objective optimization, the collaborative optimization of vehicle longitudinal and lateral control is achieved, providing a new technical means for improving intersection traffic efficiency and energy conservation and emission reduction.

[0131] Example 2:

[0132] An electronic device comprising a memory and a processor;

[0133] The memory is used to store computer programs;

[0134] The processor is configured to implement the method described in Example 1 when executing the computer program.

[0135] Example 3:

[0136] A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method described in Example 1 is implemented.

[0137] Example 4:

[0138] A computer program product includes a computer program, which implements the method described in embodiment 1 when executed by a processor.

[0139] In the above embodiments, references to "this embodiment" in the specification indicate that a particular feature, structure, or characteristic described in conjunction with the embodiment is included in at least some embodiments, but not necessarily all embodiments. Multiple occurrences of "this embodiment" do not necessarily refer to the same embodiment.

[0140] In the above embodiments, although the invention has been described in conjunction with specific embodiments thereof, many alternatives, modifications, and variations of these embodiments will be apparent to those skilled in the art based on the foregoing description. For example, other memory structures (e.g., dynamic RAM (DRAM)) may be used with the embodiments discussed. The embodiments of the present invention are intended to encompass all such alternatives, modifications, and variations that fall within the broad scope of the appended claims.

[0141] Regarding the computer-readable storage medium in this embodiment, those skilled in the art will appreciate that all or part of the steps in the aforementioned method embodiments can be implemented using hardware associated with the computer program. The aforementioned computer program can be stored in a computer-readable storage medium. When executed, the program performs the steps in the aforementioned method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0142] The electronic terminal provided in this embodiment includes a processor, a memory, a transceiver and a communication interface. The memory and the communication interface are connected to the processor and the transceiver and complete communication with each other. The memory is used to store computer programs, the communication interface is used for communication, and the processor and the transceiver are used to run computer programs so that the electronic terminal executes the various steps of the above method.

[0143] In this embodiment, the memory may include a random access memory (RAM), and may also include a non-volatile memory (non-volatile memory), such as at least one disk storage.

[0144] The above-mentioned processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, and discrete hardware components.

[0145] The present invention can be used in a wide variety of general-purpose or special-purpose computing system environments or configurations, such as personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, and distributed computing environments that include any of the above.

[0146] The present invention may be described in the general context of computer-executable instructions, such as program modules, executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, and the like that perform specific tasks or implement specific abstract data types. The present invention may also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communications network. In a distributed computing environment, program modules may be located in both local and remote computer storage media, including storage devices.

[0147] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not limiting. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention can be modified or replaced by equivalents without departing from the purpose and scope of the technical solutions, which should all be included in the scope of the claims of the present invention.

Claims

1. A predictive control method for intelligent vehicle cooperative traffic at a signalized intersection based on vehicle-road-cloud interaction, characterized by: The following steps are involved: S1: Construct a traffic scenario model for a multi-lane signalized intersection, including defining the lane positions and sequence numbers of vehicles in each lane and adjacent vehicles; S2: Establish the longitudinal tracking state model and lateral motion model of the vehicle in the scene, and form the longitudinal and lateral coupling state model of the vehicle; S3: Design an optimization problem within the MPC framework and construct a multi-objective cost function that comprehensively considers vehicle efficiency, economy, traffic feasibility, and speed incentives. S4: Set vehicle speed constraints, acceleration constraints, safe following distance constraints, and lane change collision safety constraints in the MPC optimization problem; S5: Solve the MPC optimization problem based on the receding horizon optimization, and output the speed control sequence and lane change decision of the main vehicle in the prediction horizon as the vehicle control strategy.

2. The predictive control method for intelligent vehicle cooperative traffic model at a signalized intersection based on vehicle-road-cloud interaction according to claim 1 is characterized by: In step S1, the position of the vehicle on the road is defined as (i, j), where i is the lane index of the vehicle and j is the serial number of the vehicle in the lane. The main vehicle is recorded as (m, 0), the vehicle in front of the main vehicle is recorded as (m, 1), the following vehicles closest to the main vehicle in the left and right lanes are recorded as (l, 0) and (r, 0), respectively, and the front vehicles in the left and right lanes are recorded as (l, 1) and (r, 1), respectively. The possible lane change index is described by δ(k) = {-1, 0, 1}, where -1 indicates a left lane change, +1 indicates a right lane change, and 0 indicates no lane change. Vehicles in the leftmost lane cannot change lanes to the left, and vehicles in the rightmost lane cannot change lanes to the right. The lane index i of each vehicle is recorded, where N is the total number of lanes. Right-turning vehicles do not participate in the control strategy, and only left-turning and straight-moving vehicles are considered.

3. The predictive control method for intelligent vehicle cooperative traffic model at a signalized intersection based on vehicle-road-cloud interaction according to claim 1 is characterized by: In step S2, a discrete-time dynamic model is established for the longitudinal tracking state of the vehicle. The speed of the jth CAV in the i-th lane at time t+1 is determined by its speed and acceleration at time t. The position of the jth CAV in the i-th lane at time t+1 is related to its acceleration and speed at time t. Let S, V, and a be the position, speed, and acceleration of the CAV, respectively. The relationship between them is expressed as: In (i,j),t+1 =V (i,j),t +a (i,j),t ×Δt Where V (i,j),t+1 is the speed of the j-th CAV in the i-th lane at time t+1, V (i,j),t is the speed of the j-th CAV in lane i at time t, a (i,j),t The acceleration of the jth CAV in the i-th lane at time t, Δt is the sampling time; S (i,j),t+1 is the position of the jth CAV in the i-th lane at time t+1, S (i,j),t is the position of the j-th CAV in the i-th lane at time t; The longitudinal state space formula of a single CAV is: x i,j (k+1)=A·x i,j (k)+B·u i,j (k) x i,j (k)=[S i,j (k),V i,j (k)] Τ u i,j (k)=[a i,j (k)] where x i,j is the state vector of vehicle (i, j), u i,j is the control input vector of vehicle (i, j), k is the discrete time index, and t is the time step.

4. The predictive control method for intelligent vehicle cooperative traffic model at a signalized intersection based on vehicle-road-cloud interaction according to claim 1 is characterized by: In step S2, a discrete-time dynamic model is established for the lateral motion of the vehicle, and the state variables and control inputs are: Among them, y i,j is the lateral displacement of vehicle i relative to the lane centerline, with left deviation being positive and right deviation being negative; is the lateral velocity, is the lateral acceleration; The sampling time is Δt, and the discretized model is:

5. The predictive control method for intelligent vehicle cooperative traffic model at a signalized intersection based on vehicle-road-cloud interaction according to claim 1 is characterized by: In step S2, the longitudinal and lateral models of the vehicle are coupled to construct an extended state space model of the vehicle. The overall state space is expanded to: in represents the state vector of the longitudinal motion of vehicle (i, j), represents the longitudinal velocity of vehicle (i, j); The coupled dynamic equations are expressed as follows: Among them A lon represents the longitudinal motion state matrix, A lat represents the lateral motion state matrix, B lon represents the longitudinal motion input matrix, B lat represents the lateral motion input matrix, Represents the total lateral and longitudinal control input of vehicle (i, j).

6. The predictive control method for intelligent vehicle cooperative traffic model at a signalized intersection based on vehicle-road-cloud interaction according to claim 1 is characterized by: The multi-objective cost function in step S3 includes a lane tracking term, a speed error term, a traffic feasibility term, and an acceleration term; The lane tracking item is used to make the host vehicle as close to the predetermined target lane as possible; The speed error term represents the deviation between the vehicle’s current speed and the desired speed; The traffic feasibility item introduces the average speed information of different lanes to encourage the main vehicle to change lanes to a lane with higher traffic efficiency; The acceleration term is used to measure the energy consumption caused by acceleration; Each part is multiplied by the corresponding weight coefficient and combined to form the overall objective function. Taking into account the performance and control objectives, the objective function of the optimization problem is expressed as: Where t is the current time from the start of the simulation time, k is the kth step beyond the prediction horizon, and w l ,w v ,w t ,w a is the weight factor, M represents the number of prediction steps, y(k) represents the lane where the vehicle is located at the kth step, and y ref,p (k) represents the target lane expected at step k, v m,0 (k) represents the current speed of the host vehicle at step k, represents the target speed expected by the main vehicle in lane p∈δ(k) at step k, represents the target speed expected by the main vehicle in lane p∈δ(k) at step k, It represents the maximum average target speed of different lanes at time t within the predicted horizon, represents the acceleration of the host vehicle, p represents the lane-changing behavior, and δ(k) represents the lane-changing parameter, which describes the vehicle's movement intention.

7. The predictive control method for intelligent vehicle cooperative traffic model at a signalized intersection based on vehicle-road-cloud interaction according to claim 1 is characterized by: The vehicle speed constraint, acceleration constraint, safe following distance constraint, and lane change collision safety constraint in step S4 are specifically: The vehicle's speed must not exceed the set maximum speed and must not be lower than the set minimum speed; the vehicle's acceleration must not exceed the set maximum value and must not be lower than the set minimum value; the main vehicle and the vehicle in front must maintain a headway of at least the set minimum time; during the lane change process, a safe distance must be maintained between the main vehicle and vehicles in the left and right lanes to avoid collisions. The constraints are expressed as follows: 0≤v m,0 ≤v max a min ≤a m,0 ≤a max Among them, v m,0 is the vehicle speed, v max is the maximum speed limit of the lane, a min The minimum acceleration of the main vehicle, a max is the maximum acceleration of the main vehicle; Among them, x m,1 is the position of the preceding vehicle, x m,0 For the main vehicle position, is the minimum headway, R0 is the minimum gap between the main vehicle and the preceding vehicle; (x i,j (k)-x m,0 (k)) 2 ≥δ i,j (k)·R c where x i,j (k) is the state variable of vehicle (i, j) at step k, x m,0 (k) is the state variable of the main vehicle at step k, δ i,j (k) is the lane-changing discrete action, R c is the minimum gap between the main vehicle and the preceding vehicle in different lanes, L i,j (k) represents the index of the lane where the k-th vehicle (i, j) is located, L m,0 Indicates the lane index of the host vehicle.

8. The predictive control method for intelligent vehicle cooperative traffic model at a signalized intersection based on vehicle-road-cloud interaction according to claim 1 is characterized by: In step S5, a rolling horizon optimization solution method is used to predict the future motion state of the main vehicle in the time domain by collecting traffic signal phase and timing SPaT information and real-time status information of surrounding vehicles. The cost function is minimized by the quadratic programming QP numerical optimization algorithm, and the solution is iteratively solved in the rolling horizon to obtain the optimal acceleration sequence and lane change decision of the main vehicle. Only the first control input of the optimization sequence is applied to the vehicle, and the optimization process is updated at the next moment based on the newly acquired status information. The green wave speed V of the vehicle in the green light state at time t is signal (t) is represented by the following formula: Where d(t) represents the distance between the current vehicle and the signal intersection, t rem is the remaining time of the current phase, represents the maximum speed of green wave at time t, V max Indicates the maximum allowed speed, T g and T r They are the time when the current lane opens green light and the time when the current lane opens red light in each signal cycle, k = 0, 1, 2, 3, ..., ∞ represents the acceptable green light time sequence, which will be satisfied in the specific calculation The first window of is used as the basis for the final k value; The target speed of the main vehicle at time t under p∈δ(t) is determined as a function of the headway: Among them, α1 and α2 are constants for adjusting the speed amplitude, (c, 1) means the front vehicle is driving in different lanes, l is the left lane, m is the main vehicle lane, r is the right lane, s c,1 is the position of the preceding vehicle, v c,1 is the speed of the preceding vehicle, is the reference time, i.e. the ideal headway; s m,0 (t) is the position of the main vehicle at time t, v m,0 (t) is the main vehicle speed at time t; the headway calculation formula is as follows: where s c,0 (t) is the position of the main vehicle in different lanes at time t, v c,0 (t) is the speed of the main vehicle in different lanes at time t; At time k, the target velocity on the M-step prediction horizon is expressed as: In the forecast horizon, the target expected speed Changes over time, It is the average value of the target expected speed under the lane change behavior p within the prediction horizon: is the maximum value of the target speed between different lanes: in, is the average speed in different lanes; Target acceleration a t,p (k) The target speed between two steps in each prediction interval is expressed as: By using the above parameters, the target speed on the predicted horizon is determined, and the cost function is optimized to obtain the minimum cost under different lane change operations on the predicted horizon, and then the optimal Calculate the optimal lane and get the optimal acceleration

Images