CACC controller parameter optimization method considering communication delay and related device

Abstract

The invention discloses a CACC controller parameter optimization method considering communication delay and a related device, and the method comprises the steps: building a CACC model, and determining the stability constraint and communication delay boundary of the CACC model; constructing an optimization objective function according to the CACC model, and constructing a Gaussian process model based on the optimization objective function; and based on the Gaussian process model, solving optimal parameters in the stable region by using a Bayesian optimization algorithm. On the basis of a stability optimization framework of the CACC model, the Bayesian optimization algorithm is introduced innovatively, accurate optimization of parameters of the CACC controller under communication delay constraints is realized, and queue stability and disturbance suppression capability are improved.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of intelligent transportation and vehicle control, and relates to a CACC controller parameter optimization method considering communication delay and a related device. BACKGROUND

[0002] A vehicle platoon system relies on vehicle-to-vehicle (V2V) and vehicle-to-road (V2I) communication technologies to achieve cooperative control, which can significantly improve road traffic efficiency and reduce fuel consumption. It is a core application scenario of automatic driving technology. The cooperative adaptive cruise control (CACC) system is a key technology for longitudinal control of vehicle platoons. It needs to obtain real-time information about the state of the preceding vehicle, such as acceleration and speed, through V2V communication. However, communication delays caused by factors such as multipath effects, protocol encapsulation / de-encapsulation, and high-density traffic data conflicts in the wireless communication process can reduce the stable parameter region of the CACC controller, reduce the robustness of the platoon, and even cause instability phenomena such as amplification of spacing errors and propagation of disturbances upstream along the platoon, which threatens system safety.

[0003] The existing CACC controller parameter optimization research has the following limitations: Low efficiency of optimization algorithm: Most studies use traditional heuristic algorithms such as particle swarm optimization (PSO) and genetic algorithms. A large number of iterations are required for convergence, and it is easy to fall into local optima, making it difficult to meet the real-time control requirements of vehicle platoons (control period is usually 0.1-0.5s).

[0004] Insufficient use of prior information in parameter space: Traditional algorithms do not combine the physical meaning of CACC controller parameters (such as spacing error weight which should be positive) with historical optimization data, resulting in strong blindness in parameter search, especially in dynamic communication delay scenarios, where optimization accuracy and efficiency are difficult to balance.

[0005] Lack of targetedness in parameter optimization under communication delay: Some studies do not explicitly couple the communication delay and the controller parameters, and only select parameters in the stable region through trial and error, which cannot guarantee the optimality of the parameters under different communication delay conditions.

[0006] In the prior art, the communication delay boundary has been determined through numerical analysis (such as the vehicle headway , actuator time delay , and the conservative upper limit of communication delay is 0.2s), and a CACC model stability optimization framework has been constructed. However, the traditional optimization algorithm used in this framework has the problems of low efficiency and easy falling into local optima. SUMMARY

[0007] The purpose of the present application is to provide a CACC controller parameter optimization method considering communication delay and a related device, which solves the problems of low optimization efficiency and easy falling into local optima in the prior art.

[0008] To achieve the above object, the application adopts the following technical solutions to achieve: The CACC controller parameter optimization method considering communication delay comprises: Constructing a CACC model and determining the stability constraint and communication delay boundary of the CACC model; According to the CACC model, an optimization objective function is constructed, and based on the optimization objective function, a Gaussian process model is constructed; Based on the Gaussian process model, the optimal parameters in the stable region are solved by using a Bayesian optimization algorithm.

[0009] Further, the expression of the CACC model is:

[0010] Among them, is the expected acceleration of the nth vehicle, is the time, is the weight coefficient of the spacing error, the vehicle speed, the relative speed and the front vehicle acceleration respectively, is the spacing error, is the vehicle speed, is the relative speed, is the front vehicle acceleration, is the communication delay, is the model error correction term. Further, the stability constraint comprises a first type of stable region and a second type of stable region;

[0011] The first type of stable region is: and ; The second type of stable region is: and ; Among them, , , , , is the actuator time delay, is the weight coefficient of the spacing error, the vehicle speed, the relative speed and the front vehicle acceleration respectively, is the communication delay.

[0012] Further, the optimization objective function is:

[0013] , , ,

[0014] wherein, is an optimization objective function, is the maximum spacing error of the i-th vehicle, is a stable region.

[0015] Further, the acquisition function of the Gaussian process model is:

[0016] wherein, is an optimization objective function, is the current optimal parameter, is the parameter to be sampled.

[0017] Further, the process of solving by using the Bayesian optimization algorithm includes: substitute the sampling point into the CACC model, combine the local traffic flow comprehensive trajectory, calculate the actual objective function value, and verify whether the parameter meets the stability constraint, if not, discard and reselect the sampling point; add the reselected sampling point to the sample set, update the mean function and the covariance matrix of the Gaussian process model, and improve the fitting accuracy of the Gaussian process model to the parameter space; repeat the process of “acquisition function point selection → parameter evaluation → model update”, when the number of iterations reaches the preset value, or the change amount of the objective function value of the continuous 3 iterations is less than , stop iteration, and output the current optimal parameter.

[0018] The CACC controller parameter optimization system considering communication delay includes: a first modeling module, configured to construct a CACC model, and determine the stability constraint and the communication delay boundary of the CACC model; a second modeling module, configured to construct an optimization objective function according to the CACC model, and construct a Gaussian process model based on the optimization objective function; a solving module, configured to solve the optimal parameter in the stable region by using a Bayesian optimization algorithm based on the Gaussian process model.

[0019] A computer program product includes computer instructions, which instruct a computer to execute the steps.

[0020] A terminal device includes a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the method.

[0021] A computer-readable storage medium stores a computer program, which is executed by a processor to implement the steps of the method.​

[0022] Compared with the prior art, the present application has the following beneficial effects: The present application provides a CACC controller parameter optimization method considering communication delay, which introduces a Bayesian optimization algorithm based on the stability constraint of the CACC model and the communication delay boundary, uses a Gaussian process model to mine prior information in the parameter space, and efficiently searches for optimal parameters in the stable region, solving the problem of multiple iterations and easy falling into local optimum of traditional optimization algorithms, and realizing real-time optimization of CACC controller parameters and improvement of queue stability in a communication uncertain environment. The present application innovatively introduces a Bayesian optimization algorithm based on the stability optimization framework of the CACC model, without the need to reconfigure the core framework, reducing the difficulty of engineering landing. Through the Bayesian optimization algorithm, the present application realizes precise optimization of the CACC controller parameters under the constraint of communication delay, improves the queue stability and disturbance suppression capability. The present application improves the parameter optimization efficiency and accuracy, better adapts to the vehicle queue control demand in a communication uncertain environment, and is suitable for longitudinal control of intelligent connected vehicle queue in a wireless communication uncertain environment. Compared with the traditional algorithm which is easy to fall into local optimum, the Bayesian optimization balances exploration and utilization by collecting functions, and under the condition of communication delay of 0.2s (worst case), the standard deviation of the tail vehicle spacing error is reduced to 0.022m, which is further reduced by 8.7% compared with the PSO optimization result (0.0241m). At the same time, the parameters can be dynamically adjusted in real time in combination with the communication delay, and when the communication delay of 0.1s increases to 0.2s, the Bayesian optimization can complete parameter updating within 30 iterations, adapting to the dynamic change scenario of communication delay. The Bayesian optimization uses prior information through the Gaussian process model, and the iteration number is only 50% of that of the particle swarm algorithm, and the time consumption of single optimization is <0.5s, meeting the real-time control demand of vehicle queue. BRIEF DESCRIPTION OF DRAWINGS

[0023] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiments. It should be understood that the following drawings only show some embodiments of the present application, and therefore should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can also be obtained without creative labor on the basis of these drawings.

[0024] Figure 1 The flowchart of the CACC controller parameter optimization method considering communication delay of the present application.

[0025] Figure 2 The overall framework diagram of the cooperative lane changing strategy of the present application.

[0026] Figure 3 The stable region diagram under four communication delay conditions of the present application.

[0027] Figure 4This is a diagram illustrating the process of generating a local traffic flow integrated trajectory according to the present invention.

[0028] Figure 5 The graph shows the spacing error performance of four optimal controllers when the communication delay is 0.1s.

[0029] Figure 6 The graph shows the spacing error performance of four optimal controllers when the communication delay is 0.15s.

[0030] Figure 7 This is a diagram showing the attenuation rate of disturbances in the vehicle queue for different optimal controllers under different communication delay conditions according to the present invention.

[0031] Figure 8 This is a schematic diagram of a CACC controller parameter optimization system structure considering communication delay, which is a preferred embodiment of the present invention.

[0032] Figure 9 This is a schematic diagram of the electronic device structure according to a preferred embodiment of the present invention. Detailed Implementation

[0033] The following description, in conjunction with the accompanying drawings, illustrates exemplary embodiments of this application, including various details to aid understanding. These should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of this application. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0034] Obviously, the described embodiments are only some, not all, of the embodiments in this application. All other embodiments obtained by those skilled in the art based on the embodiments in this application without inventive effort are within the scope of protection of this application.

[0035] It should be noted that the terminals involved in the embodiments of this application may include, but are not limited to, mobile phones, personal digital assistants (PDAs), wireless handheld devices, tablet computers, personal computers (PCs), MP3 players, MP4 players, wearable devices (e.g., smart glasses, smartwatches, smart bracelets), smart home devices, and other smart devices.

[0036] Furthermore, the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this article generally indicates that the preceding and following related objects have an "or" relationship.

[0037] The present invention will now be described in further detail with reference to the accompanying drawings: See Figure 1 and Figure 2 This invention provides a method for optimizing CACC controller parameters considering communication delay. Based on CACC model construction, stability constraint derivation, and communication delay boundary determination, a Bayesian optimization algorithm is introduced to solve for the parameters. Specifically, the method includes the following steps: Step 1: CACC model construction and stability constraint derivation.

[0038] CACC linear model: considering communication delay With actuator time delay The model expression is:

[0039] in, For the first The expected acceleration of the vehicle. For time, These are the weighting coefficients for spacing error, vehicle speed, relative speed, and acceleration of the vehicle in front, respectively. For spacing error, For the speed of this vehicle, For relative velocity, To accelerate the vehicle in front, Due to communication delay, This is the model error correction term. For the parameters to be optimized, the lower-level controller satisfies... ( ).

[0040] Stability constraint: Transform the time-domain model into a frequency-domain transfer function using the Laplace transform. Combining the Routh-Hurwitz criterion (local stability) with transfer function magnitude constraints (String stability), two types of stable regions are derived:

[0041]

[0042]

[0043] Type I stable region: and ; Second type of stable region: and .

[0044] Step 2: Determine the communication delay boundary.

[0045] Using the controlled variable method combined with three-dimensional stable region numerical analysis, the communication delay boundaries under different operating conditions are determined: When the front of the car is far Actuator delay Communication delay A conservative upper limit is set at 0.2s (exceeding this limit causes the first type of stable region to disappear and the second type of stable region to shrink significantly); when communication delays... Actuator delay At that time, the headway of the train The conservative lower limit is 0.8s.

[0046] In practical applications, communication latency is collected in real time through the vehicle-mounted communication module. It determines whether the value is within the stability boundary; if it exceeds it, it triggers parameter re-optimization. For example... Figure 3 As shown, this demonstrates the communication delay. At that time, controller parameters ( The stable distribution range of ( ) reflects the impact of communication delay on the stable region.

[0047] Step 3: Extraction of local traffic flow characteristics and construction of the objective function for optimization.

[0048] Local traffic flow trajectory generation: Acceleration data from 230 real vehicles were selected. Through spectrum analysis, inverse fast Fourier transform (IFFT), and fourth-order Butterworth low-pass filtering (cutoff frequency 3 rad / s), a comprehensive trajectory representing the characteristics of local traffic flow was generated as a reference trajectory for parameter optimization. Figure 4 As shown.

[0049] Optimize the objective function: using the infinite norm of the vehicle spacing error. Infinite norm of distance error with the vehicle in front To minimize the ratio, ensuring that the disturbance decays upstream in the queue, the expression is:

[0050] , , ,

[0051] in, For the first The maximum distance error between vehicles This represents the stable region of step one.

[0052] Step 4: Solving for parameters based on Bayesian optimization.

[0053] By introducing the Bayesian optimization algorithm, efficient parameter search can be achieved. The core steps are as follows: 1. Initialize the Gaussian process model: Set the parameters to be optimized. As input, optimize the objective function As output, a Gaussian process model is constructed (the kernel function is a squared exponential kernel, considering the smooth correlation between parameters), and the stable parameters are used as initial sample points to train the model to mine prior information in the parameter space.

[0054] 2. Acquisition Function Design: The Expected Improvement (EI) function is adopted as the acquisition function to balance "exploration" (the parameter uncertainty of the unsampled region) and "utilization" (the optimal parameters of the sampled region). The formula is as follows:

[0055] in, These are the current optimal parameters. The parameters to be sampled are defined; by maximizing the EI function, the next sampling point is determined, avoiding the blind search of traditional algorithms.

[0056] 3. Iterative Optimization and Convergence Judgment: In each iteration, based on the objective function value predicted by the Gaussian process model and the uncertainty, new sampling points are selected using the EI function, substituted into the CACC model to calculate the actual objective function value, and the Gaussian process model is updated; when the number of iterations reaches a preset value or the change in the objective function value is less than a threshold (…), the convergence judgment is made. When the condition is met, stop the iteration and output the optimal parameters.

[0057] 4. Parameter verification: Substitute the optimized parameters into the stability constraints ( ) and assessment of perturbation energy decay rate ( This verifies whether the parameters meet the stability requirements and the disturbance suppression effect.

[0058] The present invention will be further described in detail below through specific embodiments: Example 1: 1. Experimental Scenario Setup This embodiment takes a single-lane CACC queuing scenario on a highway as an example. The queuing consists of 6 intelligent connected vehicles. The trajectory of the lead vehicle adopts the real vehicle trajectory of the US101 section of the NGSIM dataset, with an initial speed of 30 m / s and an initial vehicle spacing strategy based on constant time distance. Settings; Initial values ​​for CACC model parameters: , , Actuator time delay The communication delays were set to 0.1s, 0.15s, and 0.2s, respectively. The values ​​of the key parameters are shown in Table 1.

[0059] Table 1. CACC Model Parameter Values

[0060] 2. Implementation Steps Step 1: Construction of CACC Model and Stability Constraints Construct a CACC control model that considers communication delay, with the following expression:

[0061] in, For the first The expected acceleration of the vehicle. For the parameters to be optimized, the time-domain model is transformed into a frequency-domain transfer function using the Laplace transform. Combining the Routh-Herwitz criterion with series stability amplitude constraints Two types of stable regions are derived:

[0062] )

[0063] Type I stable region: ; Second type of stable region: This ensures that the parameters to be optimized are searched only within a stable region.

[0064] Step 2: Real-time monitoring and assessment of communication delays and operating conditions Real-time acquisition of communication latency via onboard V2V communication modules (such as DSRC or C-V2X) The sampling frequency is 10Hz; if detected (Stable upper limit), trigger emergency parameter optimization process (prioritize queue safety); if It performs routine parameter optimization at 0.5s intervals to adapt to dynamic changes in communication latency.

[0065] Step 3: Optimize the objective function and constraint settings With the core objective of maximizing the decay of the perturbation along the upstream of the queue, the objective function is constructed as follows:

[0066] in, For the first The maximum spacing error between vehicles is subject to the following constraints: the parameters to be optimized belong to the stable region. The parameter values ​​conform to physical meaning (e.g.) (Ensure positive adjustment of spacing error) and ensure that the optimization results are both stable and effective in control.

[0067] Step 4: Parameter optimization based on Bayesian optimization algorithm 4.1 Initialize the Gaussian process model Input layer: parameters to be optimized They form a three-dimensional parameter space, which serves as the input variables for the Gaussian process model. Output layer: based on the objective function The (distance error infinity norm ratio) is used as the model output to reflect the control performance corresponding to the parameters.

[0068] Prior sample selection: Select 3 sets of initial parameters that satisfy the stability constraints (e.g., ... , The actual objective function value is substituted into the CACC model and used as the initial training sample for the Gaussian process model to avoid excessive initial prediction bias.

[0069] 4.2 Design of Acquisition Function and Selection of Sampling Points The Expected Improvement (EI) function is used as the acquisition function to balance "exploration" (uncertainty in unsampled regions of the parameter space) and "utilization" (discovered regions of superior parameters). The formula is as follows:

[0070] in, Let x be the optimal parameter for the current iteration, and let x be the parameter to be sampled. The next optimal sampling point is determined by maximizing the EI function through the numerical optimization algorithm L-BFGS, which avoids the blind search of traditional optimization algorithms and improves the efficiency of parameter exploration.

[0071] 4.3 Iterative Updates and Model Optimization Parameter evaluation: Substitute the new sampling points into the CACC model and combine them with the local traffic flow integrated trajectory to calculate the actual objective function value. At the same time, it is verified whether the parameter meets the stability constraint. If it does not meet the constraint, it is discarded and a new sampling point is selected.

[0072] Model update: Add new sampling points By adding a sample set, the mean function and covariance matrix of the Gaussian process model are updated (the squared exponential kernel function parameters are adaptively adjusted), thereby improving the fitting accuracy of the Gaussian process model to the parameter space.

[0073] Convergence Criterion: Repeat the process of "selecting points for data acquisition function → evaluating parameters → updating the model". Convergence is determined when the number of iterations reaches 50, or the change in the objective function value is less than [a certain value] in three consecutive iterations. When the time is right, stop iterating and output the current optimal parameters. .

[0074] 4.4 Validation of Optimization Results Substitute the optimal output parameters into the CACC model and verify them using both the frequency and time domains: Frequency domain: Verifying the magnitude of the transfer function For all Established to ensure string stability; Time domain: Simulate the platoon following process, calculate the standard deviation of the tail car spacing error and the disturbance energy decay rate. To verify the effect of parameter perturbation suppression. For example... Figure 5 and Figure 6 As shown, this paper demonstrates the difference in perturbation suppression performance between the present invention and traditional models (such as the Gipps model).

[0075] Step 5: Real-vehicle level simulation verification Based on the Prescan / Simulink co-simulation platform, a CACC queuing simulation system was built, including a communication delay module and a vehicle dynamics model (such as a 15-DOF model). The parameters obtained through Bayesian optimization were loaded into the controller to simulate the queuing process under different communication delay conditions. Figure 7 As shown in Table 2, the simulation results of the optimization parameters of the traditional PSO algorithm are compared to verify the advantages of the present invention in terms of optimization efficiency (number of iterations, time consumption) and control performance (spacing error, disturbance attenuation).

[0076] Table 2 Comparison of optimization parameters of traditional PSO algorithm and the effect of the present invention

[0077] As can be seen from the comparison results in Table 2, the Bayesian optimization algorithm of the present invention achieves higher parameter optimization accuracy by halving the number of iterations through Gaussian process modeling and intelligent sampling, while significantly shortening the optimization time, adapting to the real-time control requirements of vehicle queues. In particular, under the scenario of dynamic changes in communication latency, the parameter update response speed is significantly improved, ensuring the stable operation of the queue.

[0078] Example 2: This invention also provides a CACC controller parameter optimization system that takes into account communication delay, such as... Figure 8 As shown, the system includes: a first modeling module, a second modeling module, and a solution module.

[0079] The first modeling module is used to construct the CACC model and determine the stability constraints and communication delay boundaries of the CACC model. The second modeling module is used to construct an optimization objective function based on the CACC model, and to construct a Gaussian process model based on the optimization objective function. The solver module is used to solve for the optimal parameters within the stable region based on the Gaussian process model and using the Bayesian optimization algorithm.

[0080] It is understood that the CACC controller parameter optimization system considering communication delay provided by the present invention corresponds to the CACC controller parameter optimization method considering communication delay provided in the foregoing embodiments. The relevant technical features of the CACC controller parameter optimization system considering communication delay can be referred to the relevant technical features of the CACC controller parameter optimization method considering communication delay, and will not be repeated here.

[0081] Another object of the present invention is to provide an electronic device, such as... Figure 9 As shown, it includes a memory, a processor, and a computer program stored in the memory and executable on the processor, the processor performing the steps of the CACC controller parameter optimization method that takes into account communication latency.

[0082] The CACC controller parameter optimization method considering communication delay includes the following steps: Construct a CACC model and determine the stability constraints and communication delay boundaries of the CACC model; Based on the CACC model, an optimization objective function is constructed, and based on the optimization objective function, a Gaussian process model is constructed. Based on the Gaussian process model, the optimal parameters within the stable region are solved using the Bayesian optimization algorithm.

[0083] A fourth objective of this invention is to provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the CACC controller parameter optimization method considering communication delay.

[0084] The CACC controller parameter optimization method considering communication delay includes the following steps: Construct a CACC model and determine the stability constraints and communication delay boundaries of the CACC model; Based on the CACC model, an optimization objective function is constructed, and based on the optimization objective function, a Gaussian process model is constructed. Based on the Gaussian process model, the optimal parameters within the stable region are solved using the Bayesian optimization algorithm.

[0085] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0086] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0087] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0088] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0089] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A method for CACC controller parameter optimization considering communication delay, characterized in that, The application relates to a method for determining an optimal parameter in a CACC model. The method comprises the following steps: a CACC model is constructed, and stability constraints and communication delay boundaries of the CACC model are determined; an optimization objective function is constructed according to the CACC model, and a Gaussian process model is constructed based on the optimization objective function; 2. The CACC controller parameter optimization method considering communication delay according to claim 1, characterized in that, optimal parameters in a stable region are solved by using a Bayesian optimization algorithm based on the Gaussian process model. wherein, is the desired acceleration of the vehicle, is the time, is the time, are weight coefficients of the spacing error, the host vehicle speed, the relative speed, and the preceding vehicle acceleration, respectively, is the spacing error, is the host vehicle speed, is the relative speed, is the preceding vehicle acceleration, is the communication delay, is the model error correction term.

3. The CACC controller parameter optimization method considering communication delay according to claim 1, wherein, The expression of the CACC model is as follows: The first type of stable region is: and ; The second type of stable region is: and ; wherein, , , , is the actuator time delay, are the weight coefficients of the spacing error, the ego vehicle speed, the relative speed, and the preceding vehicle acceleration, respectively, is the communication delay.

4. The CACC controller parameter optimization method considering communication delay according to claim 1, wherein, The stability constraints include a first type of stable region and a second type of stable region. ， ， ， wherein, is the optimization objective function, is the maximum spacing error for the vehicle, is the stability region.

5. The CACC controller parameter optimization method considering communication delay according to claim 1, wherein, The optimization objective function is as follows: wherein, is the objective function to be optimized, is the current best parameter, is the parameter to be sampled.

6. The CACC controller parameter optimization method considering communication delay according to claim 1, wherein, The acquisition function of the Gaussian process model is as follows: The process of solving by using the Bayesian optimization algorithm comprises the following steps: a sampling point is substituted into the CACC model, an actual objective function value is calculated in combination with a local traffic flow comprehensive trajectory, and it is verified whether the parameter meets the stability constraints; if not, the sampling point is discarded, and the sampling point is reselected; The "acquisition function point selection→parameter evaluation→model update" process is repeated, and when the number of iterations reaches a preset value, or the target function value change amount of three consecutive iterations is less than , the iteration is stopped, and the current optimal parameters are output.

7. A CACC controller parameter optimization system considering communication delay, characterized in that, the reselected sampling point is added to a sample set, a mean function and a covariance matrix of the Gaussian process model are updated, and the fitting accuracy of the Gaussian process model on a parameter space is improved. The application further relates to a device for determining an optimal parameter in a CACC model. The device comprises: a first modeling module for constructing a CACC model and determining stability constraints and communication delay boundaries of the CACC model; 8. A computer program product comprising computer instructions, characterized in that, a second modeling module for constructing an optimization objective function according to the CACC model and constructing a Gaussian process model based on the optimization objective function; 9. A terminal device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, a solving module for solving optimal parameters in a stable region by using a Bayesian optimization algorithm based on the Gaussian process model.

10. A computer-readable storage medium storing a computer program, the computer program comprising instructions that, when executed by a computer, cause the computer to perform the method of any one of claims 1-9. The computer instructions instruct the computer to execute the steps of any one of claims 1-6. The processor executes the computer program to realize the steps of the method of any one of claims 1-6. The computer program is executed by the processor to realize the steps of the method of any one of claims 1-6.

Images