Near signal area mixed traffic finite time control method based on CPS

Abstract

The invention belongs to the technical field of intelligent traffic, and discloses a near signal area mixed traffic finite time control method based on a CPS, and the method comprises the following steps: S1, collecting the information of a signal lamp in a near signal area and the information of a vehicle in a mixed traffic vehicle-following scene in real time; s2, constructing a vehicle longitudinal dynamics model and a mixed sub-queue model; s3, near-signal area single-point green wave passing speed control considering time constraint is carried out; s4, designing an inter-queue finite time coupling controller; and S5, constructing a finite time event triggering switching controller under the dynamic communication topology. According to the method, a finite time coupling control strategy is provided so as to quickly eliminate the driving cut-in influence of people; the method can enable the vehicle to converge and stabilize again within a limited time, and has obvious pertinence and effectiveness for solving the problem of driver driving in a near-signal area mixed traffic stable fleet.

Description

Technical Field

[0001] The present invention belongs to the technical field of intelligent transportation, and in particular relates to a CPS-based finite-time control method for mixed traffic in a near-signal area. Background Art

[0002] Existing research on near-signal areas mostly targets homogeneous traffic, focusing on longitudinal speed guidance and lateral lane-changing strategies in lane-changing areas. The former is usually based on distributed control, with the fuel consumption, travel time, comfort, etc. of a single vehicle during driving as optimization objectives, and is converted into solutions for optimal speed, acceleration, and trajectory. The latter usually regards the lane-changing behavior of vehicles in near-signal areas as a form of interference with the fleet or group of vehicles, focusing on how to eliminate traffic conflicts between vehicles and how to reduce the negative impact of lane-changing behavior on traffic efficiency, and proposing appropriate control strategies to solve problems such as fleet instability caused by lane-changing problems. In the near-signal area scenario, CAVs in mixed traffic are used to perceive traffic signal information and make speed plans, and control strategies are designed to reduce the negative effects caused by forced lane changes, thereby addressing the two major characteristics of near-signal areas, reducing fuel consumption, and improving traffic efficiency.

[0003] The hybrid traffic system presents typical characteristics of a cyber-physical system, and its core is that the control mechanism of CAV is essentially a dynamically coupled system in which discrete information drives continuous physical processes. Based on this, this application takes the perspective of cyber-physical systems to explore how to utilize the information of vehicles, roads, and traffic signals in mixed traffic scenarios near signal areas in physical space, and through real-time analysis, integration, and processing of multi-source information in cyber space and the issuance of precise control instructions, optimize the continuous physical space-time with discrete information, and ultimately achieve a deep integration of physical space and cyber space. This analysis and solution of mixed traffic problems near signal areas based on the CPS (cyber-physical system) perspective is an important idea in new hybrid traffic scenarios.

[0004] Swarm control methods are optimized with the goal of maximizing overall benefits by coordinating the interactions between individuals. Research shows that the integration of vehicle networking and autonomous driving technologies has promoted the innovation of vehicle swarm control. The local optimization methods brought by traditional single-vehicle intelligence have obvious limitations and are difficult to cope with the current complex traffic environment. Control optimization based on vehicle queues as a group is one of the effective means of current vehicle swarm control. In mixed traffic environments, establishing a vehicle swarm based on fleets can give full play to the advantages of CAVs and achieve indirect influence on HVs, thereby making the evolution of the vehicle swarm develop in the desired direction and improving driving safety and traffic efficiency.

[0005] Existing research on finite-time control has yielded significant results in the multi-agent field, which requires both rapid convergence and stability. However, in platoon control for heterogeneous traffic systems near signal areas, unstable mixed-vehicle fleets in these areas place special demands on time as a control metric. Considering the disturbances caused by frequent lane changes, it is necessary to design control strategies that can eliminate the negative impact of these disturbances within a finite timeframe, thereby ensuring rapid convergence for the heterogeneous traffic system. Summary of the Invention

[0006] In order to solve the traffic congestion problem caused by periodic stops and stops due to signal light switching and human drivers cutting in and out, the present invention proposes a CPS-based finite-time control method for mixed traffic in near-signal areas based on the scenario of near-signal areas, considering the need to stabilize the unstable fleet and form a re-clustered stable fleet under certain time constraints, so as to quickly eliminate the impact of human drivers cutting in.

[0007] The present invention adopts the following technical solutions to solve the above problems:

[0008] The present invention provides a CPS-based finite-time control method for mixed traffic near a signal area, comprising the following steps:

[0009] S1. Real-time collection of traffic light information near the signal area and vehicle information in mixed traffic following scenarios;

[0010] Mixed traffic refers to the coexistence of human-driven HVs and intelligent connected vehicles (CAVs) on the road;

[0011] Among them, intelligent connected cars are equipped with laser radar or millimeter wave radar. Intelligent connected cars can accurately sense the speed information of the vehicle in front and follow and receive information from roadside units (RSUs). Human drivers can only perceive surrounding environmental variables through human observation.

[0012] S2. Construct a vehicle longitudinal dynamics model and a mixed traffic sub-platoon model;

[0013] S3. Single-point green wave speed control in the near-signal area considering time constraints;

[0014] S4. Design a finite-time coupling controller between queues;

[0015] S5. Construct a finite-time event-triggered switching controller under dynamic communication topology.

[0016] Furthermore, step S2 includes the following sub-steps:

[0017] S2.1 Construct a vehicle longitudinal dynamics model;

[0018]

[0019] Where Mi represents the mass of the i-th vehicle; a i represents the acceleration of the i-th vehicle; T i (t) and r i They represent the mechanical efficiency of the transmission system, braking torque and tire radius of the i-th vehicle respectively; v i represents the speed of the i-th vehicle; C, ρ and A i They represent the air resistance coefficient, air density and frontal area of ​​the i-th vehicle respectively; g is the acceleration due to gravity; f i is the drag coefficient of the i-th vehicle; θ is the road slope;

[0020] According to control theory, the position and velocity of the vehicle are expressed as the following two-dimensional nonlinear model:

[0021]

[0022] in:

[0023]

[0024] We can get:

[0025]

[0026] Where p i (t) represents the position of the i-th vehicle; ||f(p i ,v i )||≤d max ,||u i ||<τ, that is, the nonlinear function in the system and the input of the leader are bounded, where d max ,τ are all positive constants;

[0027] S2.2 builds a mixed sub-queue model;

[0028] The mixed vehicle platoon near the signal area consists of multiple mixed sub-platoons, each of which contains a CAV as the lead vehicle and several HVs as following vehicles.

[0029] Define the initial state of the subqueue:

[0030]

[0031] Where, P l 、V l 、A l 、L l Represent the position, velocity, acceleration and length of the lth sub-queue respectively; represents the position of the first CAV vehicle in the lth sub-queue; v l,i 、a l,irepresent the speed and acceleration of the i-th vehicle in the l-th sub-queue, respectively; k represents the number of HV vehicles in the l-th sub-queue; Indicates the position of the last vehicle in the lth sub-queue; len indicates the length of the vehicle.

[0032] Furthermore, step S3 includes the following sub-steps:

[0033] S3.1 Assume that a traffic light cycle is T, the phase change of the traffic light is red first and then green, and the phase of the green light is t G , the red light phase is t R , satisfying T = t G +t R ;

[0034] S3.2 When a CAV enters a near-signal area, it obtains signal phase and timing information through V2X communication and divides the cluster based on the green light time window and the maximum number of vehicles passing through.

[0035] Set a CAV vehicle as the leading vehicle in the convoy;

[0036] Assume that the leading vehicle enters the near signal area at the tth second within a traffic light cycle, 0≤t≤T; the leading vehicle has a speed of v0 and an initial position of p0, X stop Indicates the starting position of the no-lane-change zone. The distance from the leader vehicle to the traffic light is X. stop -p0; the number of following cars in the convoy is k0;

[0037] If t≤t G , the convoy enters the current cycle from the green light phase, and the conditions for the convoy to pass the traffic light before the end of the next green light phase are:

[0038] t c +t n +t≤T+t G

[0039]

[0040] Where, t c is the current travel time; t n is the required travel time; d0 represents the fixed vehicle spacing;

[0041] By adjusting v0, the convoy can pass the traffic light without stopping;

[0042] If t≥t G , then the convoy enters the current cycle from the red light phase, with:

[0043]

[0044] The following conditions must be met:

[0045]

[0046] Where, t n1 Indicates the time required to pass the traffic light; t c1 Indicates that it exceeds t c Travel time; t n * represents the required travel time; t n * When there is no cutting vehicle, it is t n , if there is an intruding vehicle, it is converted into:

[0047] t n1 =((X stop -p0)-v0(2T-tt lc )+k0d0) / v0

[0048] Where, t lc is the lane change completion time.

[0049] Furthermore, in step S4, the expression of the inter-queue finite time coupling controller is as follows:

[0050]

[0051] Where, γ1 and χ are coupling adjustment parameters, γ1>0; u represents the control output; L B represents the communication topology; β, p, φ, m, n, g, α, and h are all parameters; q represents the weighting coefficient; Q represents the coupling coefficient; s represents the sliding surface; ε2 represents the velocity error; ε1 represents the position error; F represents the nonlinear term of the following vehicle; f0 represents the nonlinear term of the leading vehicle; and u0 represents the control input of the leading vehicle.

[0052] α, β∈R + ,g,h,p,q∈N and are odd numbers and satisfy 1<p / q<2,g / h>p / q;

[0053] m, n∈N are odd numbers and satisfy 0<m / n<1.

[0054] Furthermore, in step S5, the expression for triggering the switching controller by a finite time event under the dynamic communication topology is:

[0055]

[0056] Where u origin Indicates the controller before the change; is the monotonically increasing sampling time of the i-th CAV at the time of event triggering, where Yes [T l ,T l+1) within the kth sampling moment; u new Indicates the new controller after the change; Indicates the communication topology after the change; s new represents the sliding surface after the change; Indicates the speed error after the change; Indicates the position error after the change.

[0057] Beneficial effects:

[0058] The present invention proposes a CPS-based finite-time control method for mixed traffic in near-signal areas. This method can enable vehicles to converge and stabilize again within a finite time. It is clearly targeted and effective in solving the problem of human drivers cutting in from a stable fleet in mixed traffic in near-signal areas.

[0059] A key point of the present invention is that it takes into account that vehicles in the near-signal area change lanes frequently, and each lane change will have an impact on the fleet and even the traffic conditions. And this impact will increase with the increase in the number of lane changes, so for each lane change process, its impact needs to be eliminated as soon as possible. On the one hand, eliminating the impact of the fleet under certain time constraints will reduce the vibration amplitude of the overall speed fluctuation of the fleet, thereby improving the stability of the entire fleet. On the other hand, if the cut-in behavior occurs frequently, the longer the time interval between the two cut-in behaviors, the more likely it is to avoid the "coupling amplification" effect brought about by the two cut-ins. Therefore, it is necessary to eliminate the impact of the cut-in earlier under certain time constraints.

[0060] 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

[0061] Figure 1 This is a flow chart of a CPS-based finite time control method for mixed traffic near signal areas of the present invention;

[0062] Figure 2 This is a schematic diagram of the speed control for single-point green wave traffic in a CPS fleet;

[0063] Figure 3 This is a schematic diagram of clustering based on traffic light cycles;

[0064] Figure 4 This is a schematic diagram of a queue recovering to stability after being disturbed under single in / out conditions;

[0065] Figure 5 Schematic diagram of the queue evolution under multiple cut-ins and cut-outs between coupled fleets. DETAILED DESCRIPTION

[0066] To make the technical solutions, advantages, and purposes of the present invention more clear, the technical solutions of the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the described embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0067] In this embodiment, the evolution of mixed vehicle queues near the signal area under CPS is as follows: Figure 2 shown.

[0068] like Figure 1-Figure 5 As shown, the present invention provides a CPS-based finite time control method for mixed traffic near signal areas, comprising the following steps:

[0069] S1. Real-time collection of traffic light information near the signal area and vehicle information in mixed traffic following scenarios;

[0070] In the near-signal area, vehicle driving is restricted by signals, requiring drivers to constantly pay attention to the changes in signal light status while also considering the driving conditions of vehicles in adjacent lanes. These factors make it difficult for drivers to concentrate, and vehicles are prone to frequent starting and stopping, acceleration and deceleration, and other issues. In order for vehicles in the near-signal area to reach different destinations after passing through the intersection, lane changes before the approach to the near-signal area can be considered forced lane changes. Due to the "time constraint" requirement in the near-signal area, that is, the fleet must eliminate the impact of cutting in under certain time constraints, which will reduce the vibration amplitude of the overall speed fluctuation of the fleet, thereby improving the stability of the entire fleet.

[0071] The definition of mixed traffic includes, on the one hand, human drivers operating in a free-wheeling state, and, on the other hand, intelligent connected vehicles operating as the leading vehicle or in a controlled state in the same lane; the former represents an ordinary vehicle, while the latter represents a connected vehicle equipped with intelligent systems. Human drivers perceive environmental variables through their own eyes, while intelligent vehicles are equipped with sensors such as LiDAR or millimeter-wave radar. Therefore, the automated driving system can accurately sense the speed of the preceding vehicle, follow it, and receive information from the RSU.

[0072] S2. Construct a vehicle longitudinal dynamics model and a mixed traffic sub-platoon model;

[0073] Secondly, the CAV longitudinal control model is designed: the CAV is modeled through vehicle dynamics to apply the cooperative lane change control strategy and subsequent control:

[0074] M i ai (t) = F T,i (t)-F R,i (t)

[0075] Where M i represents the mass of the i-th vehicle, a i represents the acceleration of the i-th vehicle, F T,i (t) and F R,i (t) is the traction and total resistance applied to the i-th vehicle. Using the definitions of traction and resistance, equation (0.1) can be rewritten as:

[0076]

[0077] Where, T i (t) and r i denote the mechanical efficiency of the transmission system, braking torque and tire radius of the i-th vehicle respectively; C, ρ and A i is the air resistance coefficient, air density and frontal area of ​​the i-th vehicle, g is the acceleration due to gravity, f i is the drag coefficient of the i-th vehicle, and θ is the road slope.

[0078] According to control theory, the position and velocity of the vehicle can be expressed as the following two-dimensional nonlinear model:

[0079]

[0080] Where p i (t) and v i (t) represents the position and velocity of the i-th vehicle.

[0081]

[0082] Where ||f(p i ,v i )||≤d max ,||u i ||<τ, that is, the nonlinear function in the system and the input of the leader are bounded, where d max , τ are positive constants. The above formula can be rewritten as:

[0083]

[0084] Regarding longitudinal following in mixed traffic scenarios, since HVs can only follow the vehicle ahead by manually sensing its status, they will automatically form a sub-platoon with the lead vehicle, regardless of lane changes. This sub-platoon consists of a CAV leading the way and HVs following behind. In this scenario, a mixed vehicle platoon near the signal area can be considered to consist of multiple mixed sub-platoons.

[0085] Each sub-platoon consists of a CAV as the lead vehicle and several HVs as followers. If a CAV is followed by one or more CAVs, then the CAV is an independent sub-platoon. To describe and analyze the driving behavior of a sub-platoon, the initial state of the sub-platoon is defined as follows:

[0086]

[0087] Where, P l 、V l 、A l 、L l Represent the position, velocity, acceleration and length of the lth sub-queue respectively. represents the position of the first CAV vehicle in the lth sub-queue, v l,i 、a l,i Represent the speed and acceleration of the i-th vehicle in the l-th sub-queue respectively. k represents the number of HV vehicles in the l-th sub-queue, It represents the position of the last vehicle in the lth sub-queue, and len represents the length of the vehicle.

[0088] S3. Single-point green wave speed control in the near-signal area considering time constraints;

[0089] With the goal of allowing mixed-vehicle convoys to pass through traffic lights without stopping, the following single-point green wave speed control strategy is designed using the connected autonomous vehicle perception mechanism established at the information level:

[0090] Assume that a traffic light cycle is T, the phase change of traffic lights is red light first and then green light, and the phase of green light is t G , the red light phase is t R , satisfying T = t G +t R .

[0091] When a CAV enters a near-signal area, it obtains signal phase and timing (SPaT) information through V2X communication and implements cluster division based on the green light time window and the maximum number of vehicles passing. Then, a CAV is set as the head vehicle of the cluster. Assume that the leading vehicle enters the near-signal area at the tth (0≤t≤T) second within a traffic light cycle. The leading vehicle has a speed of v0, an initial position of p0, and X stop Indicates the starting position of the no-lane-change zone. The distance from the starting point of the simulated road to the traffic light is X stop -p0. The total number of following cars in the convoy is k0.

[0092] If t≤t G , the convoy enters the cycle from the green light phase. Under this condition, the convoy passes before the end of the next green light phase:

[0093]

[0094] Where, t c is the current travel time, t n For the required travel time, the following conditions must be met:

[0095] t c +t n +t≤T+t G

[0096] Use the above formula to adjust v0 so that the team can pass the traffic light without stopping.

[0097] If t≥t G , then the convoy enters the cycle from the red light phase,

[0098]

[0099] The following conditions must be met:

[0100]

[0101] Where, t n * When there is no cutting vehicle, it is t n , if there is an intruding vehicle, it is converted into:

[0102] t n1 =((X stop -p0)-v0(2T-tt lc )+k0d0) / v0

[0103] Where, t lc is the lane change completion time.

[0104] S4. Design a finite-time coupling controller between queues;

[0105] Define the position error and speed error for the i-th sub-queue in the fleet:

[0106]

[0107] Where, represents the position error between the i-th sub-queue and the head vehicle of other sub-queues in the fleet, represents the speed error between the i-th sub-queue and the head vehicle of other sub-queues, a ij is the communication coefficient in the i-th row and j-th column of the adjacency matrix A, b i is the communication coefficient between the leader and the follower.

[0108] use Kronecker (Kronecker product) rewrites the above formula:

[0109]

[0110] Where,

[0111]

[0112] Derivative of the above formula:

[0113]

[0114] Where, F=[f(P1,V1),…,f(P n ,V n )] T ,u=[u1,u2,…,u n ] T .

[0115] The traditional non-singular terminal sliding mode surface (NTSM) converges too slowly because the exponent of the state variable x1 in the sliding mode surface is less than 1. In order to increase the absolute value of the state derivative and improve the convergence speed of the terminal sliding mode control, based on this concept and taking into account the non-singular requirement of the terminal sliding mode control, the non-singular fast terminal sliding mode surface (NFTSM) is rewritten and the sliding mode surface is defined for the model:

[0116]

[0117] Where, s=[s1,s2,…,s n ] T ,α,β∈R + ,g,h,p,q∈N and are odd numbers and satisfy 1<p / q<2,g / h>p / q to ensure the non-singularity of the sliding surface.

[0118] Derivative the sliding surface and rewrite it using Hadamard product:

[0119]

[0120] In order to eliminate the nonlinear switching term in the control law and avoid system chattering, an "attractor" is usually used to design the sliding mode control law. Among them, the terminal attractor can make the state reach the sliding surface in a finite time and has good robustness to model errors and external disturbances. Therefore, according to a terminal attractor with a negative exponential term of the state

[65] Rewrite the reaching law:

[0121]

[0122] Where φ, γ>0, m, n∈N are odd numbers and satisfy 0<m / n<1. Combining the above formulas, the non-singular fast terminal sliding mode control law can be derived as follows:

[0123]

[0124] Since 1<p / q<2, g / h>1, the exponents in the controller are all greater than zero and there are no negative exponential terms, which shows that the sliding mode control method based on NFTSM and terminal attractor avoids the singularity problem, and the control time is continuous without chattering.

[0125] Since the sliding surface defined above cannot guarantee the strong queue stability of the system, the present invention introduces a coupled sliding surface to enhance the interaction between CAVs and ensure stability. The form is as follows:

[0126]

[0127] Where, i∈V,q>0,Π i (t) and s i The relationship between (t) is

[0128] Π(t)=Qs(t)

[0129] Where Q is obtained by the following definition:

[0130] Description i and s i relationship;

[0131]

[0132] Where q is the weighting coefficient. Based on the above definition, S i and s i The relationship is described as S(t) = Qs(t), which is in the following form:

[0133]

[0134] =Although the control law designed based on the terminal attractor can give full play to the advantages of NFTSM, it cannot fully exert its advantages in the control of global state variables.

[0135] In order to solve the above problems, a terminal attractor reaching law based on the coupled sliding surface is designed in combination with the coupled sliding surface:

[0136]

[0137] When the system is close to the equilibrium point, the higher-order terms on the right side of Π are ignored. When the system is far from the equilibrium point, the higher-order terms on the right side of Π play a major role due to the coupled sliding surface. By adding a coupling adjustment parameter χ to tighten the coupled sliding surface and solve the problem of slow convergence caused by Π(t), the finite-time coupled controller is designed as follows:

[0138]

[0139] Where, γ1 and χ are coupling adjustment parameters, and γ1>0.

[0140] S5. Construct a finite-time event-triggered switching controller under a dynamic communication topology;

[0141] This step proposes an event-triggered switching control strategy to address the vehicle platoon reorganization and controller update issues caused by dynamic communication topologies. By applying different control strategies to different traffic scenarios, the advantages of CAVs can be maximized and traffic efficiency can be improved.

[0142] In addition to the single-point green wave speed planning strategy for the CAV lead vehicle, which needs to consider cutting in and out, the event-triggered switching control mechanism is mainly used in the limited-time control strategy in mixed vehicle fleets.

[0143] Assume that there is an infinite sequence of events [T l ,T l+1 ), where l=0,1,…∞, and h1>T l+1 -T l >h0>0, h0 and h1 are two positive constants. Each time interval in the sequence does not overlap, and the initial time T0=0. l ,T l+1 ) There are m samples in the sequence, and the nth CAV is at the kth sampling time Leave the charged queue, where Indicates that the CAV has not left the time period and communication is still possible; Indicates that the CAV has left the original queue. In the process, the communication is interrupted due to the departure of CAV, which changes the communication topology G. This change will continue until the next time a CAV joins or leaves the queue. Indicates the sampling time of the nth vehicle. represents the number of times the queue has changed, and Where q≥1 represents the set of all possible topologies when vehicles join or leave the queue.

[0144] From the above, we can see that in the event-triggered switching control process, is the monotonically increasing sampling time of the i-th CAV at the time of event triggering, where Yes [T l ,T l+1 ) is the kth sampling moment in the triggering time. The switching moment must be the sampling moment. The controller is updated instantly and passes through the zero-order hold until the next sampling trigger moment

[0145] For the process of the i-th CAV vehicle leaving the queue. Assume that the i-th CAV is Then the next trigger moment can be determined by satisfying the following trigger conditions

[0146]

[0147] Among them, P rear P0 and P1 represent the positions of the tail and lead CAVs, respectively. The trigger logic ensures that the distance between the tail CAV and the lead CAV exceeds the distance limit determined by the maximum number of vehicles. Indicates the vehicle is delayed The trigger logic is when Δp change >L, that is, it is triggered when the deceleration distance exceeds a single vehicle length to ensure a safe distance. i -v ref (t)| is the difference between the current speed and the expected speed of the team. The trigger logic is when the speed difference exceeds the threshold ∈ v It is triggered in real time to avoid speed fluctuations that may undermine the stability of the fleet.

[0148] Similarly, for the process of the i-th CAV vehicle joining the queue. Assume that the i-th CAV is at the time of the last event triggering Then the next trigger moment can be determined by satisfying the following trigger conditions

[0149]

[0150] in Indicates that the added CAV is the new lead vehicle.

[0151] For vehicles in the original platoon, their sub-platoons still follow the finite-time control law, but the topology changes caused by vehicles joining / leaving the platoon lead to changes in the controller. The following describes the changes in the platoon using the departure of a CAV.

[0152] Assume that the CAV in the i-th sub-queue is in the communication topology G i The set of neighbors in When the topology switches to G due to the departure of the i+1th CAV i+1 When , the neighbor set is updated to The process of defining the adjacency matrix and the Laplace matrix is ​​as follows: the state matrix before triggering is:

[0153]

[0154] When the CAV leaves the queue, the state matrix becomes:

[0155]

[0156] The updating rules of the sliding surface and the controller after topology switching are:

[0157]

[0158] Where, For topology G i+1 The new neighbor set in is the new adjacency matrix weight, is the new traction matrix weight. i+1 The adjacency matrix A i+1 , update the weights:

[0159]

[0160] The update rules for position, velocity error function and sliding surface are as follows:

[0161]

[0162] The finite time controller is updated to:

[0163]

[0164] Therefore, the control rule update for the distributed finite-time sub-queues in the fleet is as follows:

[0165]

[0166] For the CAV vehicle leaving, the control change process of the vehicle is:

[0167]

[0168] When applying event-triggered switching control, Zeno behavior (i.e., an infinite number of triggers within a finite timeframe) should be avoided. In this invention, since the discrete sampling period of CAV vehicle sensors is fixed at 0.1s, the number of sampling times within a finite timeframe is limited, and therefore the number of event triggers is also limited. This lower bound on the event interval based on the event triggering condition prevents Zeno behavior.

[0169] It is hereby stated 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 may 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 CPS-based finite-time control method for mixed traffic near signal areas, characterized in that: The following steps are involved: S1. Real-time collection of traffic light information near the signal area and vehicle information in mixed traffic following scenarios; Mixed traffic refers to the coexistence of human-driven HVs and intelligent connected vehicles (CAVs) on the road; Among them, intelligent connected cars are equipped with laser radar or millimeter wave radar. Intelligent connected cars can accurately sense the speed information of the vehicle in front and follow and receive information from roadside units (RSUs). Human drivers can only perceive surrounding environmental variables through human observation. S2. Construct a vehicle longitudinal dynamics model and a mixed traffic sub-platoon model; S3. Single-point green wave speed control in the near-signal area considering time constraints; S4. Design a finite-time coupling controller between queues; S5. Construct a finite-time event-triggered switching controller under dynamic communication topology.

2. The CPS-based finite time control method for mixed traffic near signal areas according to claim 1 is characterized in that: The step S2 includes the following sub-steps: S2.1 Construct a vehicle longitudinal dynamics model; Where M i represents the mass of the i-th vehicle; a i represents the acceleration of the i-th vehicle; T i (t) and r i They represent the mechanical efficiency of the transmission system, braking torque and tire radius of the i-th vehicle respectively; v i represents the speed of the i-th vehicle; C, ρ and A i They represent the air resistance coefficient, air density and frontal area of ​​the i-th vehicle respectively; g is the acceleration due to gravity; f i is the drag coefficient of the i-th vehicle; θ is the road slope; According to control theory, the position and velocity of the vehicle are expressed as the following two-dimensional nonlinear model: in: We can get: Where p i (t) represents the position of the i-th vehicle; ||f(p i ,v i )||≤d max ,||u i ||<τ, that is, the nonlinear function in the system and the input of the leader are bounded, where d max ,τ are all positive constants; S2.2 builds a mixed sub-queue model; The mixed vehicle platoon near the signal area consists of multiple mixed sub-platoons, each of which contains a CAV as the lead vehicle and several HVs as following vehicles. Define the initial state of the subqueue: Where, P l 、V l 、A l , L l Represent the position, velocity, acceleration and length of the lth sub-queue respectively; represents the position of the first CAV vehicle in the lth sub-queue; v l,i 、a l,i represent the speed and acceleration of the i-th vehicle in the l-th sub-queue, respectively; k represents the number of HV vehicles in the l-th sub-queue; Indicates the position of the last vehicle in the lth sub-queue; len indicates the length of the vehicle.

3. The CPS-based finite time control method for mixed traffic near signal areas according to claim 2 is characterized in that: The step S3 includes the following sub-steps: S3.1 Assume that a traffic light cycle is T, the phase change of the traffic light is red light first and then green light, and the phase of the green light is t G , the red light phase is t R , satisfying T = t G +t R ; S3.2 When a CAV enters a near-signal area, it obtains signal phase and timing information through V2X communication and divides the cluster based on the green light time window and the maximum number of vehicles passing through. Set a CAV vehicle as the leading vehicle in the convoy; Assume that the leading vehicle enters the near signal area at the tth second within a traffic light cycle, 0≤t≤T; the leading vehicle has a speed of v0 and an initial position of p0, X stop Indicates the starting position of the no-lane-change zone. The distance from the leader vehicle to the traffic light is X. stop -p0; the number of following cars in the convoy is k0; If t≤t G , the convoy enters the current cycle from the green light phase, and the conditions for the convoy to pass the traffic light before the end of the next green light phase are: t c +t n +t≤T+t G Where, t c is the current travel time; t n is the required travel time; d0 represents the fixed vehicle spacing; By adjusting v0, the convoy can pass the traffic light without stopping; If t≥t G , then the convoy enters the current cycle from the red light phase, with: The following conditions must be met: Where, t n1 Indicates the time required to pass the traffic light; t c1 Indicates that it exceeds t c travel time; t n * Indicates the required travel time; t n * When there is no cutting vehicle, it is t n , if there is an intruding vehicle, it is converted into: t n1 =((X stop -p0)-v0(2T-t-t lc )+k0d0) / v0 Where, t lc is the lane change completion time.

4. The CPS-based finite time control method for mixed traffic near signal areas according to claim 3 is characterized in that: In step S4, the expression of the inter-queue finite time coupling controller is as follows: Where, γ1,χ are coupling adjustment parameters, γ1>0; u represents the control output; L B represents the communication topology; β, p, φ, m, n, g, α, and h are all parameters; q represents the weighting coefficient; Q represents the coupling coefficient; s represents the sliding surface; ε2 represents the velocity error; ε1 represents the position error; F represents the nonlinear term of the following vehicle; f0 represents the nonlinear term of the leading vehicle; and u0 represents the control input of the leading vehicle. α, β∈R + ,g,h,p,q∈N and are odd numbers and satisfy 1<p / q<2,g / h>p / q; m, n∈N are odd numbers and satisfy 0<m / n<1.

5. The CPS-based finite time control method for mixed traffic near signal areas according to claim 4 is characterized in that: In step S5, the expression for triggering the switching controller by a finite time event under a dynamic communication topology is: Where u origin Indicates the controller before the change; is the monotonically increasing sampling time of the i-th CAV at the time of event triggering, where Yes [T l ,T l+1 ) within the kth sampling moment; u new Indicates the new controller after the change; Indicates the communication topology after the change; s new represents the sliding surface after the change; Indicates the speed error after the change; Indicates the position error after the change.

Images