Event-triggered distributed vehicle cooperative control method

Abstract

The invention discloses an event-triggered distributed vehicle cooperative control method, and relates to the technical field of networked vehicle system control. The method comprises the following steps: constructing a networked vehicle system mathematical model composed of networked vehicles in bidirectional communication interconnection; establishing a nonlinear tracking error model based on the established mathematical model of the networked vehicle system; designing an event triggering distributed model predictive control strategy for predictive time domain self-adaptive adjustment; and the recursive feasibility of the designed control strategy and the stability of the closed-loop system are verified. According to the vehicle cooperative control method, the overall stability, the cooperative consistency, the physical safety and the anti-interference performance of the network connection vehicle system are effectively guaranteed.

Description

Technical Field

[0001] This invention belongs to the field of connected vehicle system control technology, and particularly relates to an event-triggered distributed vehicle cooperative control method. Background Technology

[0002] With the acceleration of urbanization, problems such as traffic congestion, frequent accidents, and environmental pollution have become increasingly prominent. Against this backdrop, intelligent transportation systems (ITS) have emerged as a key technological solution to address urban traffic difficulties. Among them, connected vehicles (V2V), as an important component of ITS, have become a research hotspot in the global automotive and transportation sectors in recent years. In terms of traffic safety, V2V can achieve time-to-time perception beyond line of sight through vehicle-to-vehicle and vehicle-to-infrastructure communication, providing early warnings of potential collision risks. In terms of traffic efficiency, connected collaborative control can optimize traffic flow, reducing unnecessary acceleration, deceleration, and congestion. In terms of the environment and society, V2V can effectively reduce carbon emissions through optimized hypothetical strategies.

[0003] As a core research direction for connected vehicles, connected vehicle control strategies have received widespread attention and in-depth research from academia and industry in recent years. With the rapid development of related technologies, current research on connected vehicle control strategies has formed a multi-layered technical system, mainly including single-vehicle intelligent control strategies, multi-vehicle collaborative control strategies, and special scenario control strategies. In the future, connected vehicle control strategies will develop towards greater intelligence, greater collaboration, and greater safety.

[0004] Current research on connected vehicle control strategies has rapidly evolved from early single-vehicle intelligence to multi-agent collaboration and vehicle-road-cloud integration. Distributed optimization and robust control of connected vehicles is one of the core research directions in multi-agent collaborative control, aiming to address the challenges of real-time performance, safety, and adaptability in multi-vehicle collaborative decision-making under dynamic traffic environments, and significant progress has been made. However, bottlenecks such as computational efficiency, resource utilization, and real-time performance still need to be overcome.

[0005] Therefore, this invention proposes an event-triggered distributed model predictive control method for predictive time-domain adaptive adjustment of connected vehicles with bidirectional communication interconnection. By integrating a robust safety constraint tightening strategy, a predictive time-domain adaptive adjustment mechanism, and an event-triggered communication scheme, this method effectively solves the problems of excessive communication resource consumption and excessive computational burden in traditional distributed control while ensuring the overall stability, coordination consistency, physical security, and anti-interference performance of the connected vehicle system. It significantly improves the overall resource utilization efficiency of the connected vehicle system, has theoretical significance, strong feasibility, and is conducive to improving economic benefits. Summary of the Invention

[0006] The purpose of this invention is to provide an event-triggered distributed vehicle cooperative control method to solve the problems of excessive communication resource consumption and excessive computational burden in the traditional distributed control of connected vehicle systems mentioned in the background art.

[0007] To achieve the above objectives, the present invention employs the following technical solution:

[0008] This invention proposes an event-triggered distributed vehicle cooperative control method, comprising the following steps:

[0009] S1. Construct a mathematical model of a connected vehicle system composed of connected vehicles interconnected by two-way communication;

[0010] S2. Based on the constructed mathematical model of the connected vehicle system, establish its nonlinear tracking error model;

[0011] S3. Design an event-triggered distributed model predictive control strategy for predictive time-domain adaptive adjustment;

[0012] S4. Verify the recursive feasibility and closed-loop system stability of the designed event-triggered distributed model predictive control strategy with predictive time-domain adaptive adjustment.

[0013] Preferably, S1 is specifically as follows:

[0014] A connected vehicle system consisting of three interconnected vehicles communicating in both directions is constructed, and its dynamic subsystem is described as follows:

[0015]

[0016] Where, ω i (t) represents the position of the i-th connected vehicle. Let u represent the speed of the i-th connected vehicle. i (t) is the control input, w i (t) represents the input generated by external environmental interference and system noise, M represents the mass of the connected vehicle body, k represents the spring constant, and h represents the damping constant;

[0017] Define state variables Control input u i (t), then the state equation of the entire connected vehicle system is:

[0018]

[0019] in, and These are the system's state and control inputs, respectively. Let be a disturbance term, and satisfy d i (t)∈D:={d i (t): ||d i (t)||≤ξ}, where It is a known constant.

[0020] Preferably, step S2 is as follows:

[0021] Let the desired position and speed of the connected vehicle be ω. i，ref (t) = 0 and The tracking error is then:

[0022]

[0023] Define the tracking error system state variable as z i (t)=[z i1 (t), z i2 (t)] T and control input is v i (t), then the connected vehicle tracking error model is expressed as:

[0024]

[0025] Among them, the connected vehicle system i has state constraints. and input constraints in and All are convex compact sets, and In addition, the function and For a known constant It satisfies the Lipschitz condition, that is:

[0026]

[0027] in, and

[0028] Preferably, the event-triggered distributed model predictive control strategy for time-domain adaptive adjustment in S3 is as follows:

[0029] When the connected vehicle system is not in the terminal domain, distributed rolling time-domain optimization control is adopted. By integrating robust safety constraint tightening strategy, predictive time-domain adaptive adjustment mechanism and event triggering mechanism, the triggering frequency of the controller is reduced. After the connected vehicle system enters the terminal domain, the system switches to local controller and continues to use event triggering mechanism.

[0030] Preferably, step S3 is as follows:

[0031] S3.1 Determine the optimal control problem of the connected vehicle system and obtain the nominal optimal predicted trajectory;

[0032] S3.2 Robust Tightening Constraints; Tightening constraints based on the deviation between the actual state trajectory and the nominal optimal predicted trajectory;

[0033] S3.3 Design an adaptive prediction time domain update strategy; As the connected vehicle system approaches the terminal domain, design a prediction time domain that is gradually shortened to maintain the feasibility of the connected vehicle system;

[0034] S3.4, Dual-mode event triggering mechanism design; realizes the switching to the local controller after the connected vehicle system enters the terminal domain.

[0035] Preferably, step S3.1 is as follows:

[0036] Define the prediction time domain and the trigger time series as follows: and in Execution time is expressed as And the shrinkage in the prediction time domain is expressed as

[0037] Connected vehicle system i The optimal control problem at time t is expressed as:

[0038]

[0039]

[0040] in, and These represent the feasible input trajectory and the corresponding predicted state trajectory, respectively, with the superscript *optimal case; For terminal state constraints, and These are state and control input constraints, respectively. Let represent the hypothetical state trajectory of the neighboring connected vehicle systems of the i-th connected vehicle system, where Represents the set of all neighboring systems of connected vehicle system i;

[0041] Will Defined as the neighboring connected vehicle system j at the current trigger time The latest communication time point was in Under these circumstances, the corresponding assumed control input trajectory Defined as:

[0042]

[0043] Otherwise, regarding the situation The corresponding assumption control input trajectory Defined as:

[0044]

[0045] Coupling cost function Recorded as Its expression is:

[0046]

[0047] The self-stage cost and the collaborative stage cost are respectively expressed as: and For the terminal cost; Q i Q ij R i P i >0 is a weighted matrix with appropriate dimensions; Cost function Recorded as

[0048] The control input v of the connected vehicle system i i (t) is represented as:

[0049]

[0050] Hypothetical problem At the initial moment It can be solved.

[0051] Preferably, step S3.2 is as follows:

[0052] In the prediction time domain The actual state trajectory within and the nominal optimal predicted state trajectory Due to the actual state trajectory Deviation from the nominal optimal predicted trajectory

[0053]

[0054] The tightening state constraints are estimated using the Minkowski set subtraction method, specifically as follows:

[0055]

[0056] in,

[0057] For satisfying the state constraints and control input constraints nominal system In local controller Under control, there exists a robust positive invariant set. have and Where matrix Ξ i >0.

[0058] Preferably, S3.3 is as follows:

[0059] In the present At the trigger point, the shortest prediction time range that guarantees the feasibility of the optimal control problem is:

[0060]

[0061] Subsequently, the time-domain scale of the shrinking prediction Designed as:

[0062]

[0063] Where, 0 < μ i <1 is an adjustment parameter to ensure feasibility; The trigger parameter is defined and the condition is met. get The predicted time domain will show a decreasing or non-increasing trend.

[0064] Preferably, step S3.4 is as follows:

[0065] The status of the connected vehicle system enters the terminal domain Ω i (ε i When ), the controller switches to local controller K. i z i (t); Local controller K i z i (t) An event-triggered mechanism is adopted, and the controller remains unchanged between two triggering moments, i.e. Therefore, the state trajectory The following dynamic equations are satisfied:

[0066]

[0067] Define measurement error as The dual-mode trigger condition is obtained:

[0068]

[0069] in, As the trigger threshold, Define the trigger parameters to be designed; 0<α i γ i <1 represents the tuning parameter;

[0070] The next trigger time is:

[0071]

[0072] in,

[0073] When the connected vehicle system state has not entered the terminal domain, the lower and upper bounds of the controller's execution time are: The designed event triggering mechanism will not cause the Zeno phenomenon;

[0074] For closed-loop systems When the connected vehicle system enters the terminal domain Ω i (ε i Within a given timeframe, based on the trigger time, the lower limit of the execution time for consecutive trigger points is:

[0075]

[0076] Where ι3:=ι1+ι2||K i ||.

[0077] Preferably, step S4 is as follows:

[0078] Verify control input It is an optimization control problem A feasible solution is found, ensuring the recursive feasibility of the proposed control strategy;

[0079] Verify that once the connected vehicle system enters the terminal domain, it remains within the terminal domain, thus verifying the stability of the closed-loop system.

[0080] Compared with the prior art, the beneficial effects of the present invention are:

[0081] (1) In this invention, the vehicle cooperative control method employs distributed rolling time-domain optimization control when the system state has not entered the terminal domain. By integrating a robust safety constraint tightening strategy, a predictive time-domain adaptive adjustment mechanism, and an event-triggered communication scheme, the triggering frequency of the controller is effectively reduced. After the system state enters the terminal domain, the method switches to a local controller and continues to use the event-triggered mechanism, significantly reducing computational complexity. This vehicle cooperative control method effectively ensures the overall stability, cooperative consistency, physical security, and anti-interference performance of the connected vehicle system.

[0082] (2) The present invention verifies through simulation that the control strategy can effectively solve the problems of excessive communication resource consumption and excessive computing burden in traditional distributed control while ensuring the overall stability, coordination consistency, physical security and anti-interference performance of the connected vehicle system, and significantly improve the overall resource utilization efficiency of the connected vehicle system. Attached Figure Description

[0083] Figure 1 This is a flowchart of the event-triggered distributed vehicle cooperative control method in this invention;

[0084] Figure 2 This is a structural block diagram of the connected vehicle system composed of three interconnected vehicles with bidirectional communication in this invention;

[0085] Figure 3 This is a schematic diagram showing the position and speed trajectories of three interconnected, bidirectional communication vehicle systems in this invention;

[0086] Figure 4 This is a schematic diagram of the control input trajectory of three interconnected connected vehicle systems with bidirectional communication under the vehicle cooperative control method of the present invention;

[0087] Figure 5 This is a schematic diagram illustrating the triggering time of the control controller in the vehicle cooperative control method of the present invention;

[0088] Figure 6 This is a schematic diagram illustrating the prediction of the time-domain evolution process in the vehicle cooperative control method of this invention. Detailed Implementation

[0089] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0090] Example 1:

[0091] See Figure 1 The event-triggered distributed vehicle cooperative control method includes the following steps:

[0092] S1. Construct a mathematical model of the system consisting of three interconnected vehicles with bidirectional communication.

[0093] Specifically, the mathematical model of the system consisting of three interconnected vehicles with bidirectional communication is established as follows:

[0094] Consider a system consisting of three interconnected vehicles communicating in two directions. Its dynamic subsystem is described as follows:

[0095]

[0096] Where, ω i (t) represents the location of the i-th connected vehicle system. Let u represent the speed of the i-th connected vehicle system. i (t) is the control input, w i (t) represents the interference input, M represents the mass of the connected vehicle, k represents the spring constant, and h represents the damping constant;

[0097] For ease of analysis, state variables are defined. Control input u i (t), then the state equation of the entire system is:

[0098]

[0099] Where, x i (0)=x i0 , and These are the system's state and control inputs, respectively. Let be a disturbance term, and satisfy d i (t)∈D:={d i (t): ||d i (t)||≤ξ}, where It is a known constant.

[0100] S2. Based on the constructed mathematical model of the connected vehicle system, establish its nonlinear tracking error model.

[0101] Specifically, the nonlinear tracking error model for the connected vehicle system is as follows:

[0102] Assume the desired position and speed of the connected vehicle are ω. i，ref (t) = 0 and The tracking error is defined as:

[0103]

[0104] Define the tracking error system state variable z i (t)=[z i1 (t), z i2 (t)] T and control input v i (t), then the tracking error model of the connected vehicle system can be expressed as:

[0105]

[0106] Among them, the connected vehicle system i has state constraints. and input constraints in and All are convex compact sets, and In addition, the function and For a known constant It satisfies the Lipschitz condition, that is:

[0107]

[0108] in and

[0109] S3. Design an event-triggered distributed model predictive control strategy for predictive time-domain adaptive adjustment, specifically including the following steps:

[0110] S3.1 Description of the optimization control problem:

[0111] Define the prediction time domain and the trigger time series as follows: and in In addition, the execution time is expressed as And the shrinkage in the prediction time domain is expressed as Connected vehicle system i The optimal control problem at time t is expressed as:

[0112]

[0113]

[0114] in, and These represent the feasible input trajectory and the corresponding predicted state trajectory, respectively, with the superscript *optimal case; For terminal state constraints, and These are state and control input constraints, respectively; in addition, Let represent the hypothetical state trajectory of the neighboring connected vehicle systems of the i-th connected vehicle system, where Represents the set of all neighboring systems of connected vehicle system i;

[0115] Will Defined as the neighboring connected vehicle system j at the current trigger time The latest communication time point was in Under these circumstances, the corresponding assumed control input trajectory Defined as:

[0116]

[0117] Otherwise, regarding the situation The corresponding assumption control input trajectory Defined as:

[0118]

[0119] Coupling cost function Abbreviated as Its expression is:

[0120]

[0121] The self-stage cost and the collaborative stage cost are respectively expressed as: and For the terminal cost; Q i Q ij R i P i >0 is a weighted matrix with appropriate dimensions; for simplicity, regarding Cost function Recorded as The control input v of the connected vehicle system i i (t) can be expressed as:

[0122]

[0123] Hypothetical problem At the initial moment It can be solved.

[0124] S3.2 Robust Tightening Constraints:

[0125] Consider in the prediction time domain The actual state trajectory within and the nominal optimal predicted state trajectory They are obtained from the following formulas:

[0126]

[0127] As can be seen from (17) and (18), the actual state trajectory affected by the disturbance It usually deviates from the nominal optimal predicted trajectory

[0128] Lemma 1: At the same triggering time Prediction time domain is The upper bound of the state prediction bias at time is:

[0129]

[0130] in,

[0131] Proof: Subtracting the actual trajectory from the nominal optimal predicted trajectory, and applying the triangle inequality and the Lipschitz condition, we can obtain:

[0132]

[0133] Next, using the Gronwall-Bellman inequality, we can obtain:

[0134]

[0135] Therefore, by using Minkowski set subtraction, we can establish... To estimate the compaction constraints; for ease of stability analysis, relatively conservative compaction constraints were adopted:

[0136]

[0137] in,

[0138] For satisfying the state constraints and control input constraints nominal system In local controller Under control, there exists a robust positive invariant set. have and Where matrix Ξ i >0.

[0139] S3.3 Adaptive Prediction Time-Domain Update Strategy:

[0140] Typically, as the system approaches the terminal domain, a shorter prediction time can maintain system feasibility; therefore, a gradually shortening prediction time is designed. Firstly, in the current... At the trigger point, the shortest prediction time range that guarantees the feasibility of the optimal control problem is:

[0141]

[0142] Subsequently, the time-domain scale of the shrinking prediction Designed as:

[0143]

[0144] Where, 0 < μ i <1 is an adjustment parameter to ensure feasibility; The trigger parameter is defined and the condition is met. Therefore, according to (24), we can obtain The predicted time domain will show a decreasing or non-increasing trend, such as To simplify calculations, the initial prediction time domain is set to the same value for all connected vehicles.

[0145] S3.4, Dual-mode event triggering mechanism design:

[0146] Due to nonlinear functions It has input affine properties, that is, it satisfies and It is stable;

[0147] Lemma 2: For states constrained and control input constraints nominal system There exists a local state feedback controller Condition 1) is met It is the positive invariant set of the nominal system. in This lemma can be viewed as a special case, in which as well as

[0148] The dual-mode scheme indicates that the system state enters the terminal domain Ω. i (ε i When ), the controller switches to the local state feedback controller K. i z i (t); Since there is no optimization solution for this process, there is no need to consider the transmission between connected vehicles; In addition, in order to further save computing resources, the local controller K i z i (t) An event-triggered mechanism is adopted, and the controller remains unchanged between two triggering moments, i.e. Therefore, the state trajectory The following dynamic equations are satisfied:

[0149]

[0150] Define measurement error as The following are the conditions for obtaining dual-mode triggering:

[0151]

[0152] in, As the trigger threshold, Define the trigger parameters to be designed; 0<α i γ i <1 represents the tuning parameter;

[0153] The next trigger time is:

[0154]

[0155] in, Therefore, when the system state has not entered the terminal domain, the lower and upper bounds of the controller's execution time are: The designed event triggering mechanism will not cause the Zeno phenomenon;

[0156] Lemma 3: To ensure that the dual-mode triggering mechanism does not exhibit the Zeno phenomenon after the system enters the terminal domain, i.e., all triggering times are non-overlapping; for closed-loop systems When the system state enters the terminal domain Ω i (ε iWithin ) , the trigger time is determined by (27), then the lower limit of the execution time for consecutive trigger time points is:

[0157]

[0158] Where ι3:=ι1+ι2||K i ||.

[0159] Proof: For We can obtain:

[0160]

[0161] By comparing the principles, we can conclude that:

[0162]

[0163] The following can be obtained:

[0164]

[0165] By rearranging the above inequalities, we can see that:

[0166]

[0167] Verification complete.

[0168] S4. Verify the recursive feasibility and closed-loop system stability of the proposed control strategy, specifically including the following steps:

[0169] S4.1 Recursive Feasibility Analysis:

[0170] Lemma 4: Given and And simultaneously satisfy and It can be obtained

[0171] Proof: For achievable Define auxiliary variable h := z b -z a +τ, we can get because It can be known and then according to It can be known This indicates

[0172] Theorem 1: For a closed-loop connected vehicle system (4), the recursive feasibility of the proposed control strategy can be guaranteed if the following conditions are met: i) ii) iii)

[0173] Proof: Consider the following control input:

[0174]

[0175] In the time interval In the middle, according to (26), we get:

[0176]

[0177] Applying the Gronwall-Bellman inequality to (34) yields:

[0178]

[0179] Will Substituting (35), and combining condition i) and the triangle inequality, we get:

[0180]

[0181] In addition, within the time interval Within, according to the comparison principle, conditions ii) and (36), we obtain:

[0182]

[0183] This gives us the state. Meet terminal constraints;

[0184] Next, regarding the time interval The candidate control input (33) obviously satisfies (10); for the time interval Based on (36), it can be deduced that This yields the candidate control input. Satisfy control input constraints

[0185] according to In the time interval Inside, it is known that there are In the time interval within, from and It can be obtained This verifies that the state trajectory satisfies the state tightening constraint.

[0186] Therefore, control input It is an optimization control problem A feasible solution is found, and the recursive feasibility of the proposed control strategy is guaranteed.

[0187] S4.2, Stability analysis of closed-loop system:

[0188] Definition 2: Under the condition that recursive feasibility is satisfied, the closed-loop connected vehicle system (4) under adaptive time-domain event-triggered distributed model predictive control remains stable if the following conditions are met:

[0189]

[0190] in,

[0191]

[0192] In addition, if the conditions are met Where 0 < α i <1, and This means that once the closed-loop system enters the terminal domain, it will remain within the terminal domain.

[0193] The system state will converge to set Π i It is given by the following formula:

[0194]

[0195] in,

[0196]

[0197] Proof: The verification of this theorem can be discussed in two cases.

[0198] First, consider Let the optimal cost function be denoted as... It can be obtained satisfy:

[0199]

[0200] According to (36), we can obtain:

[0201]

[0202] Substituting (33) and (44) into (43), we get:

[0203] in,

[0204] For ΔV i 1 Using Holder's inequality, we can obtain:

[0205]

[0206] In (20) and (35), by using and It can be observed that:

[0207]

[0208] For the second component ΔV i 2 By using formula (35) and We can obtain:

[0209]

[0210] For the third component ΔV i 3 According to the triangle inequality, we know that:

[0211]

[0212] For the fourth component ΔV i 4 We can obtain:

[0213]

[0214] Based on (20) and (35), we can obtain:

[0215]

[0216] According to (36), we can obtain:

[0217]

[0218] Next, regarding the situation We can obtain:

[0219]

[0220] Regarding the situation We can obtain:

[0221]

[0222] The above two cases (57) and (58) can be summarized as follows:

[0223]

[0224] Combining the above (55), (56) and (59), we can obtain:

[0225]

[0226] The sum of the four components (51)-(53) and (60) yields the result; therefore, it is concluded that the state in the system will enter the terminal domain Ω in a finite time. i (ε i);

[0227] Consider when z i0 ∈Ω i (ε i When selecting For Lyapunov functions, we can obtain:

[0228]

[0229] Will Substituting (25) into (61), we get:

[0230]

[0231] Next, based on the design process of the dual-mode event triggering mechanism, we can obtain:

[0232]

[0233] Based on the triggering conditions:

[0234]

[0235] Therefore, Ω can be obtained. i (ε i ) is an invariant domain, meaning that once the system state enters the terminal domain, it remains within the terminal domain; summarizing the above two cases, the stability verification of the closed-loop system is complete.

[0236] Experimental verification:

[0237] To verify the adaptive predictive time-domain event-triggered distributed model predictive control strategy for connected vehicle systems provided in this embodiment, simulation experiments were conducted using MATLAB, and detailed explanations are provided below:

[0238] The system mathematical model provided in this embodiment consists of three interconnected vehicles with bidirectional communication, as follows: Figure 2 As shown, an adaptive predictive time-domain event-triggered distributed model predictive control strategy is designed. While ensuring the overall stability, coordination consistency, physical security, and anti-interference performance of the connected vehicle system, it effectively solves the problems of excessive communication resource consumption and excessive computational burden in traditional distributed control, and significantly improves the overall resource utilization efficiency of the connected vehicle system.

[0239] The parameters of the connected vehicle system are set as follows: the mass of the connected vehicle is M = 1.5 kg, and the position and speed of the connected vehicle satisfy |x i1 (t)|≤1m and|x i2 (t)|≤1m / s, and the control input satisfies |u i(t)|≤2N; spring constant k=0.25N / m, damping constant h=1.1Ns / m; assuming the disturbance is d1(t)=0.001sin(3t+1)+0.001cos(t)N, d2(t)=0.001cos(2t+1)+0.001sin(t)N, and d3(t)=0.0015sin(t+2)N; the calculated Lipschitz constant is ι1=1.64; sampling time δ and simulation time T s The values ​​were set to 0.05s and 15s respectively; the control objective was to keep the connected vehicles with disrupted two-way communication stable.

[0240] Considering the input emission characteristics of system (1), the parameter environment is designed as follows based on Lemma 2, Theorem 1 and Theorem 2: Q i = [23, 0; 0, 23], R i =5, P i =[35.2976,14.3196;14.3196,17.2256], Q ij = [0.02, 0; 0, 0.02], T0 = ​​1s, μ i =0.5, α i =0.,γ i =0.2, r i =2.4, ε i =2.2119; In addition, the initial state of the three connected vehicle systems is set as follows: x 10 = [0.6; 0], x 20 = [0.7; 0], and x 30 = [0.65; 0].

[0241] Based on the above parameters, the control method proposed in this invention was simulated and verified. The corresponding simulation results are as follows: Figures 3-6 As shown. Among them, Figure 3 The location and speed trajectories of three interconnected vehicle systems with two-way communication were demonstrated; Figure 4 This represents the control input trajectory of three bidirectional communication Internet-connected vehicle systems under the adaptive predictive time-domain event-triggered distributed model predictive control method; Figure 5 This demonstrates the triggering time of an adaptive predictive time-domain event-triggered distributed model predictive control controller. Figure 6 This illustrates the prediction time-domain evolution process of the adaptive prediction time-domain event-triggered distributed model predictive control method.

[0242] The above analysis verifies the effectiveness of the adaptive predictive time-domain event-triggered distributed model predictive control method for connected vehicle systems provided in this embodiment. When the system state has not entered the terminal domain, distributed rolling time-domain optimization control is adopted. By integrating a robust safety constraint tightening strategy, a predictive time-domain adaptive adjustment mechanism, and an event-triggered communication scheme, the triggering frequency of the controller is effectively reduced. After the system state enters the terminal domain, the system switches to a local controller and continues to use the event-triggered mechanism, significantly reducing computational complexity. Simulation verification shows that this control strategy effectively solves the problems of excessive communication resource consumption and excessive computational burden in traditional distributed control while ensuring the overall stability, coordination consistency, physical security, and anti-interference performance of the connected vehicle system, significantly improving the overall resource utilization efficiency of the connected vehicle system.

[0243] The above description is only for the purpose of helping to understand the method and core essence of the present invention, but the scope of protection of the present invention is not limited thereto. For those skilled in the art, any equivalent substitutions or modifications made to the technical solution and inventive concept disclosed in the present invention within the scope of the technology disclosed in the present invention should be covered within the scope of protection of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. An event-triggered distributed vehicle cooperative control method, characterized in that, The method comprises the following steps: S1, constructing a mathematical model of a connected vehicle system composed of connected vehicles interconnected by bidirectional communication; S2, establishing a nonlinear tracking error model based on the constructed mathematical model of the connected vehicle system; S3, designing a prediction time domain self-adaptive adjustment event-triggered distributed model predictive control strategy; S4, verifying the recursive feasibility and closed-loop system stability of the designed prediction time domain self-adaptive adjustment event-triggered distributed model predictive control strategy.

2. The event-triggered distributed vehicle cooperative control method according to claim 1, characterized in that, The S1 is specifically as follows: A connected vehicle system composed of three connected vehicles interconnected by bidirectional communication is constructed, and the dynamics subsystems are described as follows: where ω i (t) represents the position of the ith connected vehicle, u i (t) represents the speed of the ith connected vehicle, u i (t) is the control input, w i (t) is the input generated by external environmental disturbances and system noise, M is the mass of the connected vehicle body, k is the spring coefficient, and h is the damping coefficient. Definition of state variables Control input u i (t), the state equation of the entire connected vehicle system is: where and are the states and control inputs of the system, respectively, is a disturbance term, and satisfies d i (t) ∈ D := {d i (t) : ||d i (t) || ≤ ξ}, where is a known constant.

3. The event-triggered distributed vehicle cooperative control method according to claim 2, characterized in that, The S2 is specifically as follows: The desired position and velocity of the platooning vehicle are ω i，ref (t) = 0 and The tracking error is then: Define the tracking error system state variable as z i (t) = [z i1 (t), z i2 (t)] T and the control input as v i (t), then the connected vehicle tracking error model is represented as: Among them, the connected vehicle system i has state constraints. and input constraints in and All are convex compact sets, and In addition, the function and For a known constant It satisfies the Lipschitz condition, that is: wherein and 4. The event-triggered distributed vehicle cooperative control method according to claim 1 or 3, characterized in that, The prediction time domain self-adaptive adjustment event-triggered distributed model predictive control strategy in the S3 is specifically as follows: When the state of the connected vehicle system does not enter the terminal domain, a distributed rolling time domain optimization control is adopted, and through the fusion of a robust safety constraint tightening strategy, a prediction time domain self-adaptive adjustment mechanism and an event-triggered mechanism, the triggering frequency of the controller is reduced; after the state of the connected vehicle system enters the terminal domain, a local controller is switched to, and the event-triggered mechanism is continued to use.

5. The event-triggered distributed vehicle cooperative control method according to claim 4, characterized in that, The S3 is specifically as follows: S3.1, determining the optimal control problem of the connected vehicle system to obtain a nominal optimal prediction trajectory; S3.2, robust tightening constraint; based on the deviation of the actual state trajectory and the nominal optimal prediction trajectory, the constraint is tightened; S3.3, designing an adaptive prediction time domain update strategy; when the connected vehicle system approaches the terminal domain, a gradually shortened prediction time domain is designed to maintain the feasibility of the connected vehicle system; S3.4, double-mode event-triggered mechanism design; switching to a local controller after the state of the connected vehicle system enters the terminal domain is realized.

6. The event-triggered distributed vehicle cooperative control method according to claim 5, characterized in that, The S3.1 is specifically as follows: The prediction horizon and the sequence of trigger times are defined as and where The execution time is expressed as and the shrinkage of the prediction horizon is expressed as The net-connected vehicle system i at The optimal control problem at time instant is represented as where, and denote feasible input trajectories and corresponding predicted state trajectories, respectively, with the superscript * optimal case; is the terminal state constraint, and are the state and control input constraints, respectively; denotes the assumed state trajectory of the neighboring connected vehicle systems of the ith connected vehicle system, where denotes the set of all neighboring systems of the connected vehicle system i; Control input v of the connected car system i i (t) is represented as: Assumption problem At initial time Solveable.

7. The event-triggered distributed vehicle cooperative control method according to claim 6, characterized in that, The S3.2 is specifically as follows: In the prediction horizon the actual state trajectory and the nominal optimal predicted state trajectory due to the actual state trajectory deviating from the nominal optimal predicted trajectory using Minkowski set subtraction, estimate the tightened state constraints, in particular: wherein For satisfying the state constraints and control input constraints nominal system In local controller Under control, there exists a robust positive invariant set. have and Where matrix Ξ i >0.

8. The event-triggered distributed vehicle cooperative control method according to claim 7, characterized in that, The S3.3 is specifically as follows: At present The shortest prediction horizon range that guarantees the feasibility of the optimal control problem at the triggering time instant is: Subsequently, the contraction type prediction time domain scale is designed to: where 0 < μ < 1 i <1 is an adjustment parameter to ensure feasibility; is a defined trigger parameter and satisfies the condition is obtained The prediction horizon will exhibit a decreasing or non-increasing trend.

9. The event-triggered distributed vehicle cooperative control method according to claim 8, characterized in that, The S3.4 is specifically as follows: The state of the connected vehicle system enters the terminal domain Ω i (ε i ), the controller switches to the local controller K i z i (t) ; the local controller K i z i (t) adopts an event-triggered mechanism, and the controller remains unchanged between two triggering instants, i.e. Therefore, the state trajectory satisfies the following dynamics equation: The measurement error is defined as The dual-mode trigger condition is obtained: wherein, is a triggering threshold, is a triggering parameter to be designed; defined 0 < α i , γ i <1 is a tuning parameter; The next triggering time is obtained as follows: wherein The lower bound and upper bound of the execution time of the controller when the networked vehicle system state is not in the terminal domain are The designed event-triggered mechanism does not have Zeno phenomenon. For closed-loop systems When the connected vehicle system state enters the terminal domain Ω i (ε i ) inside, based on the trigger time, the lower limit of the execution time of the continuous trigger time point is: wherein i3 := i1 + i2 || K i ||.

Images