Modeling method for unified and collaborative driving model of heterogeneous vehicles in tunnel entrance area

Abstract

The invention belongs to the technical field of intelligent traffic, and discloses a modeling method of a tunnel entrance area heterogeneous vehicle unified cooperative driving model, which comprises the following steps: S1, constructing a human driving dynamics model considering tunnel environment disturbance; s2, designing a dynamic weight distribution mechanism based on communication quality perception; s3, establishing a driving model of the network-connected automatic driving vehicle in the tunnel entrance area in combination with a dynamic weight mechanism; s4, introducing a communication degradation mechanism, and establishing an automatic driving vehicle model; and S5, integrating the human driving vehicle model, the network-connected automatic driving vehicle model and the automatic driving vehicle model, and constructing a heterogeneous vehicle unified cooperative driving model. According to the method, the problem of collaborative driving modeling caused by severe environment disturbance, communication quality fluctuation and high vehicle heterogeneity in the tunnel entrance area is solved, and the accuracy, robustness and practicability of the model are remarkably improved.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of intelligent transportation, and particularly relates to a unified modeling method for cooperative driving of heterogeneous vehicles (HV, CAV, and AV) at the entrance of a tunnel, which can simultaneously consider environmental physical disturbance, communication quality fluctuation of Internet of Vehicles, and heterogeneity among Connected and Automated Vehicle (CAV), Human-driven Vehicle (HV), and Autonomous Vehicle (AV). BACKGROUND

[0002] The entrance of a tunnel is a complex traffic scene where an open road transitions to a semi-closed space. There are sudden changes in light, road line shrinkage, and significant speed limit changes. Meanwhile, the quality of communication signals is affected by wall obstruction and multipath effects, resulting in a decline in quality. It is a typical traffic bottleneck and accident-prone area. Existing research mainly focuses on two directions: one is to analyze the influence of tunnel environmental disturbance on driving behavior from the perspective of vehicle behavior modeling, and the other is to study the cooperative mechanism of CAV from the perspective of communication and control. However, the former often ignores the influence of communication fluctuation on cooperative control, and the latter often assumes stable channels and ideal communication, which cannot reflect the actual complexity in the tunnel environment. Some existing patents involve tunnel speed limit control, lighting adjustment, or vehicle speed guidance systems based on V2X, but they mainly focus on single functions and have not yet achieved unified modeling of environmental disturbance, communication quality, and vehicle heterogeneity.

[0003] In summary, the existing technology generally has the following problems: (1) lack of a model that can simultaneously reflect the comprehensive influence of environmental disturbance and communication fluctuation; (2) lack of an information fusion and weight distribution mechanism that can dynamically adjust with communication quality changes; (3) lack of a unified framework suitable for cooperative driving of heterogeneous vehicles (HV, CAV, and AV) at the entrance of a tunnel.

[0004] Therefore, there is an urgent need for a modeling method that integrates tunnel environmental disturbance, communication quality changes, and vehicle heterogeneity characteristics to improve the accuracy and robustness of the system. SUMMARY

[0005] The present application aims to overcome the shortcomings of existing technology and provide a modeling method for a unified cooperative driving model of heterogeneous vehicles at the entrance of a tunnel. This method can accurately describe the behavior characteristics of HV, CAV, and AV, and enhance the robustness and adaptability of the system in a real tunnel environment by introducing a dynamic mechanism and unified framework based on communication awareness.

[0006] To achieve the above-mentioned purpose, the present application adopts the following technical solutions:

[0007] A modeling method of a tunnel entrance area heterogeneous vehicle unified cooperative driving model, comprising the following steps:

[0008] S1. Construct a human-driven vehicle dynamics model considering tunnel environment disturbance;

[0009] S2. Design a dynamic weight distribution mechanism based on communication quality perception;

[0010] S3. Establish a tunnel entrance area connected automatic driving vehicle driving model combined with the dynamic weight mechanism;

[0011] S4. Introduce a communication degradation mechanism to establish an automatic driving vehicle model;

[0012] S5. Combine the human-driven vehicle dynamics model, the connected automatic driving vehicle driving model and the automatic driving vehicle model to construct a heterogeneous vehicle unified cooperative driving model.

[0013] Further, the step S1 comprises the following sub-steps:

[0014] S1.1 Calculate the self-driving force received by the human-driven vehicle during driving;

[0015]

[0016] In the formula, F n (t) represents the self-driving force received by the nth human-driven vehicle at time t, and the positive direction is from outside the tunnel to inside the tunnel; m n represents the mass of the nth vehicle; a n represents the maximum acceleration expected by the vehicle; v n (t) represents the speed of the nth vehicle at time t; v d represents the expected speed value of the road; and δ is an adjustment index;

[0017] S1.2 Calculate the repulsion force between human-driven vehicles;

[0018]

[0019]

[0020]

[0021]

[0022] In the formula, F n (t) represents the repulsion force received by the nth human-driven vehicle at time t, and the positive direction is from outside the tunnel to inside the tunnel; d n represents the actual inter-vehicle distance of the nth vehicle and the preceding vehicle; d n represents the expected safe inter-vehicle distance of the nth vehicle and the preceding vehicle; is the expected vehicle headway; l is the position of the nth vehicle at time t; l is the length of the vehicle body; γ is the minimum headway; b is the desired deceleration of the vehicle; γ is the adjustment index; It is the speed difference between the nth car and the (n-1)th car;

[0023] S1.3 Calculate the potential force of the tunnel, i.e., the tunnel virtual force;

[0024]

[0025]

[0026]

[0027] In the formula, This represents the virtual force exerted on the nth vehicle by the driver at time t, with the positive direction pointing from outside the tunnel to inside. It is a weighted estimate of environmental disturbance based on speed limit gradient, rate of change of light intensity and linear index; σ is the spatial attenuation coefficient; d is the distance from the tunnel entrance, that is, the distance between the current position of the vehicle and the tunnel entrance, where d=0 at the entrance; μ is the distribution center parameter; This is the speed limit inside the tunnel;

[0028] S1.4 Construct a dynamic model of human-vehicle driving in the tunnel entrance area, that is, characterize the dynamic behavior of human-vehicle driving in the tunnel entrance area by the self-driving force, the repulsive force between vehicles, and the tunnel virtual force during the driving process.

[0029] The acceleration model of the nth vehicle in the longitudinal direction at the tunnel entrance area is as follows:

[0030]

[0031] In the formula, This represents the longitudinal acceleration of the nth vehicle driven by the driver.

[0032] Furthermore, the specific content of step S2 is as follows:

[0033] For CAVs, a dynamic weight allocation mechanism for their interaction with preceding vehicle information is designed. This dynamic weight allocation mechanism dynamically calculates the fusion weight of each preceding vehicle's information based on real-time communication quality indicators, relative motion state, and distance information. The calculation formula is as follows:

[0034]

[0035] In the formula, This represents the normalized weight of the k-th vehicle at the current moment; k is the preceding vehicle number. The fusion coefficient has a range of values. This reflects the ratio of the current state to the historical weights; The basic weights of the k-th vehicle in the initial topology are based on the weights of the discretized signal-to-noise ratio (SNR) and packet loss rate (PLR) in the CPT table, reflecting the fundamental impact of communication quality on information reliability. This is the distance decay factor, reflecting the impact of the distance to the vehicle in front on the weight. For speed correction factor, It is the current speed of the vehicle. It is the speed of the kth car in front. The maximum speed limit for roads is used to reflect the impact of relative speed on communication quality; The number of valid preceding vehicles; This represents the normalized weight of the k-th vehicle at the previous moment.

[0036] Furthermore, step S3 includes the following sub-steps:

[0037] S3.1 Calculate the self-driving force experienced by the connected autonomous vehicle (CAV) during driving;

[0038]

[0039] In the formula, Let represent the self-driving force experienced by the i-th connected autonomous vehicle at time t, with the positive direction pointing from outside the tunnel to inside the tunnel; Indicate the mass of the i-th vehicle; This represents the speed of the i-th vehicle at time t;

[0040] S3.2 Calculate the repulsive force in the connected autonomous driving workshop;

[0041]

[0042]

[0043]

[0044]

[0045] In the formula, Let represent the repulsive force experienced by the i-th connected autonomous vehicle at time t, with the positive direction pointing from outside the tunnel to inside the tunnel; This represents the actual distance between the i-th vehicle and the vehicle in front; This represents the expected safe distance between the i-th vehicle and the vehicle in front; It is the position of the i-th vehicle at time t; It is the speed difference between the i-th car and the (i-1)-th car;

[0046] S3.3 Calculation of multi-vehicle cooperative feedback terms;

[0047]

[0048] In the formula, For multi-vehicle collaborative feedback items; The dynamic weight corresponding to the kth preceding vehicle is obtained through step S2; σ represents the total delay caused by information transmission and processing; σ is the adjustment parameter for the speed difference of the preceding vehicle.

[0049] S3.3 Establish a driving model for connected autonomous vehicles in the tunnel entrance area;

[0050] That is, the acceleration model of the i-th connected autonomous vehicle:

[0051]

[0052] In the formula, Let represent the acceleration of the i-th vehicle.

[0053] Furthermore, the specific content of step S4 is as follows:

[0054] Define communication status indicator variables When the signal-to-noise ratio (SNR) of a connected autonomous vehicle (CAV) is lower than a threshold or the packet loss rate (PLR) is higher than a threshold, the CAV's communication function is deemed to be faulty, and the setting is... The vehicle degenerates into an autonomous vehicle (AV) that relies solely on onboard sensors;

[0055] The AV employs a radar-based adaptive cruise control strategy, i.e., an autonomous vehicle model, whose control inputs are... for:

[0056]

[0057]

[0058] in, , To control the gain; Desired vehicle spacing; This represents the desired time interval.

[0059] Furthermore, the specific content of step S5 is as follows:

[0060] Integrate the human-driven vehicle dynamics model, the connected autonomous vehicle driving model, and the autonomous vehicle model established in steps S1-S4, and introduce vehicle type indicator variables. and communication status indicator variables A unified acceleration model is constructed, namely, a unified cooperative driving model for heterogeneous vehicles, with the following expression:

[0061]

[0062] In the formula, For vehicle type indicator variables, =1 indicates CAV / AV, =0 indicates HV; This is the self-driving force term, related to the vehicle's current speed. Related, propelling the vehicle forward; The term represents the repulsive force between vehicles, based on the distance between vehicles. Generates information to describe the interactions between vehicles.

[0063] Beneficial effects:

[0064] 1. High model accuracy and clear physical meaning: By introducing the "tunnel virtual force" term, the comprehensive impact of multiple environmental disturbances (sudden changes in illumination, velocity limiting gradient) at the tunnel entrance on HV behavior is quantified within a unified framework, which improves the model's accuracy and physical interpretability in real-world scenarios.

[0065] 2. Strong communication sensing capability and significantly improved robustness: An innovative dynamic weight allocation mechanism based on SNR / PLR real-time information is designed, enabling CAV to sense the channel status and adaptively adjust the multi-vehicle information fusion strategy, which greatly enhances the robustness and collaborative reliability of the system in the complex communication fluctuation environment of the tunnel, and effectively suppresses control instability caused by information delay or packet loss.

[0066] 3. Complete mechanism and strong practicality: Through communication status indicator variables, the mode conversion process of CAV degrading to AV after communication failure is clearly described, and the ACC control strategy is integrated, which enables the model to cope with the extreme case of communication link interruption, thus improving the practicality and completeness of the model.

[0067] 4. Unified framework, breaking through the limitations of homogenization: By using vehicle type and communication status indicator variables, a unified cooperative driving modeling framework that can accurately describe the behavior of three types of vehicles, namely HV, CAV and AV, is constructed. This solves the problem that it is difficult to systematically characterize the cooperative behavior of heterogeneous traffic flows under multiple disturbances, and provides an accurate model foundation for the design of cooperative control strategies for mixed traffic flows.

[0068] Other advantages, objectives, and features of the invention will be set forth in part in the description which follows, and in part will be apparent to those skilled in the art from the following examination, or may be learned from practice of the invention. The objectives and other advantages of the invention can be realized and obtained through the following description. Attached Figure Description

[0069] Figure 1 This is a flowchart illustrating a modeling method for a unified cooperative driving model of heterogeneous vehicles in a tunnel entrance area according to the present invention.

[0070] Figure 2 A schematic diagram of the framework for a unified cooperative driving model for heterogeneous vehicles in the tunnel entrance area;

[0071] Figure 3 This is a schematic diagram of the vehicle communication topology under three different communication scenarios;

[0072] Figure 4 A comparative diagram of different weight allocation methods;

[0073] Figure 5 Scatter plots of vehicle spatiotemporal trajectories under different CAV penetration rates;

[0074] Figure 6 A comparison chart of performance index distribution under different communication quality scenarios;

[0075] Figure 7 Heatmaps showing the spatial distribution of vehicle speeds under different communication quality scenarios. Detailed Implementation

[0076] To make the technical solutions, advantages, and objectives of the present invention clearer, 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, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the described embodiments of the present invention without creative effort are within the protection scope of this application.

[0077] like Figure 1 As shown, this invention provides a modeling method for a unified cooperative driving model of heterogeneous vehicles in a tunnel entrance area, comprising the following steps:

[0078] S1. Construct a dynamic model of human-vehicle driving considering tunnel environmental disturbances;

[0079] This step aims to quantify the impact of special environmental factors such as abrupt changes in illumination and speed limits in the tunnel entrance area on HV behavior.

[0080] S1.1 Calculate the self-driving force experienced by the driver during the driving process;

[0081]

[0082] In the formula, This represents the self-driving force experienced by the nth vehicle at time t, with the positive direction pointing from outside the tunnel to inside. This represents the mass of the nth vehicle; This represents the vehicle's desired maximum acceleration. This represents the speed of the nth vehicle at time t; δ represents the desired speed of the road; δ is the adjustment index.

[0083] S1.2 Calculate the repulsive force between the driver and the vehicle;

[0084]

[0085]

[0086]

[0087]

[0088] In the formula, This represents the repulsive force experienced by the nth vehicle at time t, with the positive direction pointing from outside the tunnel to inside. This represents the actual distance between the nth vehicle and the vehicle in front. This represents the expected safe distance between the nth vehicle and the vehicle in front. This is the expected headway. l is the position of the nth vehicle at time t; l is the length of the vehicle body; γ is the minimum headway; b is the desired deceleration of the vehicle; γ is the adjustment index; It is the speed difference between the nth car and the (n-1)th car;

[0089] S1.3 Calculate the potential force of the tunnel, i.e., the tunnel virtual force;

[0090]

[0091]

[0092]

[0093] In the formula, This represents the virtual force exerted on the nth vehicle by the driver at time t, with the positive direction pointing from outside the tunnel to inside. It is a weighted estimate of environmental disturbance based on speed limit gradient, rate of change of light intensity and linear index; σ is the spatial attenuation coefficient; d is the distance from the tunnel entrance, that is, the distance between the current position of the vehicle and the tunnel entrance, where d=0 at the entrance; μ is the distribution center parameter; This is the speed limit inside the tunnel; It is a spatial decay term based on a Gaussian function, and the simulated effect weakens as the distance increases; It is an exponential speed correction term that characterizes the intensity of the difference between vehicle speed and tunnel speed limit.

[0094] S1.4 Construct a dynamic model of human-vehicle driving in the tunnel entrance area; that is, characterize the dynamic behavior of human-vehicle driving in the tunnel entrance area by the self-driving force, the repulsive force between vehicles, and the tunnel virtual force during vehicle driving.

[0095] The acceleration model of the nth vehicle in the longitudinal direction at the tunnel entrance area is as follows:

[0096]

[0097] In the formula, This represents the longitudinal acceleration of the nth vehicle driven by the driver.

[0098] S2. Design a dynamic weight allocation mechanism based on communication quality awareness;

[0099] This step is crucial for addressing communication quality fluctuations within the tunnel. For the CAV (Communication Availability Vehicle), dynamic weights are designed for its information exchange with the preceding vehicle. The calculation of these weights comprehensively considers real-time channel conditions (SNR, PLR) and the relative distance to the preceding vehicle (using distance attenuation factors). (reflected) and relative speed difference (through speed correction factor) (Implication). During implementation, a conditional probability table (CPT) needs to be pre-established to map the quantized values ​​of SNR and PLR to a basic weight. Ultimately, the normalized weights at the current moment. The weights are obtained by smoothing the current and historical weights through a fusion coefficient ρ to ensure the stability of weight changes.

[0100]

[0101] In the formula, This represents the normalized weight of the k-th vehicle at the current moment; k is the preceding vehicle number. The fusion coefficient has a range of values. This reflects the ratio of the current state to the historical weights; The basic weights of the k-th vehicle in the initial topology are based on the weights of the discretized signal-to-noise ratio (SNR) and packet loss rate (PLR) in the CPT table, reflecting the fundamental impact of communication quality on information reliability. This is the distance decay factor, reflecting the impact of the distance to the vehicle in front on the weight. For speed correction factor, It is the current speed of the vehicle. It is the speed of the kth car in front. The maximum speed limit for roads is used to reflect the impact of relative speed on communication quality; The number of valid preceding vehicles; This represents the normalized weight of the k-th vehicle at the previous moment.

[0102] S3. Establish a driving model for connected autonomous vehicles in the tunnel entrance area by combining a dynamic weighting mechanism;

[0103] This step translates the mechanism of step S2 into the specific control behavior of the CAV. The longitudinal dynamics model of the CAV is also built on the basis of the social force model, but its core feature is that it uses V2X communication to obtain information from multiple preceding vehicles and fuses it using the dynamic weights calculated in step S2.

[0104] S3.1 Calculate the self-driving force experienced by the connected autonomous vehicle (CAV) during driving;

[0105]

[0106] In the formula, Let represent the self-driving force experienced by the i-th connected autonomous vehicle at time t, with the positive direction pointing from outside the tunnel to inside the tunnel; Indicate the mass of the i-th vehicle; This represents the speed of the i-th vehicle at time t;

[0107] S3.2 Calculate the repulsive force in the connected autonomous driving workshop;

[0108]

[0109]

[0110]

[0111]

[0112] In the formula, Let represent the repulsive force experienced by the i-th connected autonomous vehicle at time t, with the positive direction pointing from outside the tunnel to inside the tunnel; This represents the actual distance between the i-th vehicle and the vehicle in front; This represents the expected safe distance between the i-th vehicle and the vehicle in front; It is the position of the i-th vehicle at time t; It is the speed difference between the i-th car and the (i-1)-th car;

[0113] S3.3 Calculation of multi-vehicle cooperative feedback terms;

[0114]

[0115] In the formula, For multi-vehicle collaborative feedback items; The dynamic weight corresponding to the kth preceding vehicle is obtained through step S2; σ represents the total delay caused by information transmission and processing; σ is the adjustment parameter for the speed difference of the preceding vehicle.

[0116] S3.3 Establish a driving model for connected autonomous vehicles in the tunnel entrance area;

[0117] That is, the acceleration model of the i-th connected autonomous vehicle:

[0118]

[0119] In the formula, Let represent the acceleration of the i-th vehicle.

[0120] S4. Introduce a communication degradation mechanism and establish an autonomous vehicle model;

[0121] Define communication status indicator variables When the signal-to-noise ratio (SNR) of a connected autonomous vehicle (CAV) is lower than a threshold or the packet loss rate (PLR) is higher than a threshold, the CAV's communication function is deemed to be faulty, and the setting is... The vehicle degenerates into an autonomous vehicle (AV) that relies solely on onboard sensors;

[0122] This step adds a failure protection mechanism to the model to handle the extreme case of communication link interruption. This is achieved by defining a binary communication state indicator variable. To achieve this, a communication quality threshold is set for the CAV (e.g., PLR > 20%). When the communication quality is consistently below the threshold, the CAV communication function is considered to be faulty, and the setting is reset. The vehicle will then degenerate into an autonomous vehicle (AV) that relies solely on onboard sensors. The AV's control strategy will then be converted to a classic linear adaptive cruise control (ACC) model:

[0123]

[0124]

[0125] in, , To control the gain; Desired vehicle spacing; This represents the desired time interval.

[0126] S5. Construct a unified collaborative driving model for heterogeneous vehicles;

[0127] This step integrates and summarizes the previous four steps, aiming to create a universal framework capable of describing the behavior of HV, CAV, and AV simultaneously. This is achieved by introducing a vehicle type indicator variable. (1 represents CAV / AV, 0 represents HV) and communication status indicator variables To construct a unified acceleration model:

[0128]

[0129] In the formula, For vehicle type indicator variables, =1 indicates CAV / AV, =0 indicates HV; This is the self-driving force term, related to the vehicle's current speed. Related, propelling the vehicle forward; The term represents the repulsive force between vehicles, based on the distance between vehicles. Generates information to describe the interactions between vehicles.

[0130] The model judges and The value is automatically selected for each vehicle in the fleet to perform calculations, thereby achieving a unified and adaptive characterization of heterogeneous mixed traffic flow.

[0131] like Figure 2 As shown, this is a schematic diagram of the framework of the unified cooperative driving model for heterogeneous vehicles in the tunnel entrance area described in this invention. The framework clearly demonstrates how the above five steps (S1 to S5) are organically integrated, and indicates the logical relationships and information flow between each module.

[0132] In this embodiment, a simulation environment was built on the Matlab / Simulink platform to verify the effectiveness of the proposed model. The simulation road segment was set from -2000 meters upstream to 200 meters downstream, with the tunnel entrance located at coordinate 0 meters. For a mixed vehicle fleet composed of HV, CAV, and AV, different CAV penetration rates (0%, 20%, 40%) and different communication quality scenarios (high-quality HQ, medium-quality MQ, low-quality LQ, discrete degradation, and cluster degradation) were set.

[0133] To implement the method of this invention, the model parameters must first be preset. The specific values ​​of some core parameters are shown in Table 1.

[0134] Table 1. Relevant parameters of the model

[0135]

[0136] like Figure 3The diagram shown illustrates the communication topology between vehicles in three different vehicle fleets within this invention. Figure 3 (a) illustrates the communication topology of a pure CAV queue. Figure 3 (b) Demonstrates the hybrid communication topology when HV and CAV are used together. Figure 3 (c) illustrates the degraded communication topology where, after communication degradation, some CAVs degenerate into AVs that can only sense the vehicle in front. This set of figures clearly illustrates the information interaction relationships described by the model of this invention.

[0137] like Figure 4 The diagram shows a comparison of different weight allocation methods. The dynamic weight allocation mechanism proposed in this invention is compared with methods such as uniform distribution, exponential decay, and Gaussian distribution. It can be seen that the weight distribution of this method is more physically intuitive and can adaptively adjust according to communication quality.

[0138] During simulation runtime, for each vehicle in the convoy, it is identified according to its type. and communication status identifier The appropriate dynamic model is selected for state updates. For HV, its self-driving force, the repulsive force from the vehicle in front, and the tunnel virtual force are calculated in real time, and its acceleration is updated accordingly. For CAV, in addition to calculating the aforementioned social forces, the communication quality indicators (SNR, PLR), relative distance, and speed difference between it and up to three vehicles in front are obtained in real time through the V2X communication module. The fusion weights of each vehicle's information are then updated according to the dynamic weight calculation formula proposed in this invention. And calculate multiple preceding vehicle coordination terms. The total acceleration of the CAV is then obtained. At each simulation step, the communication link status of each CAV is continuously monitored; if its packet loss rate consistently exceeds a threshold... If so, its communication function is deemed to be faulty, and the setting is... The vehicle will then operate as an AV, employing a radar-based adaptive cruise control (ACC) strategy.

[0139] like Figure 5 As shown, this is a scatter plot of vehicle spatiotemporal trajectories under different CAV penetration rates. Comparing the trajectory plots with a penetration rate of ρ=0 (pure HV scenario) and ρ=0.4, it is clear that as the CAV penetration rate increases, the stability of the vehicle fleet significantly improves, the volatility of vehicle trajectories decreases, and congestion at the tunnel entrance is effectively alleviated. This demonstrates the effectiveness of the unified model of this invention and its beneficial effect on improving the stability of mixed traffic flows.

[0140] like Figure 6As shown, this is a comparison chart of performance index distribution under different communication quality scenarios. The average travel time, fuel consumption, CO2 emissions, and comfort indexes are compared under three scenarios: high-quality, medium-quality, and low-quality communication. The results show that performance deteriorates when communication quality decreases, indicating that a decline in communication quality reduces the collaborative capability of CAVs. Furthermore, the comparison with the fixed-weight mechanism shows that the dynamic weight mechanism of this invention outperforms the fixed-weight mechanism in all indicators.

[0141] like Figure 7 As shown, it is a heat map of the spatial distribution of vehicle speed under different communication quality scenarios. Figure 7 (a) In a high-quality communication scenario, it can be seen that the vehicle speed at the tunnel entrance is coordinated, fast and uniform. Figure 7 (b) In low-quality communication scenarios, the speed coordination process slows down, the gradient weakens, and the queue spacing increases. This figure visually reveals the key impact of communication quality on the effectiveness of collaborative control. Figure 6 The quantitative indicators corroborate each other's conclusions.

[0142] The stability of the unified model was analyzed using the Lyapunov-Krasovskii functional method. The theoretical derivation results show that, under certain CAV penetration and communication quality conditions, the system can achieve input-to-state stability (ISS). This provides a theoretical basis for the selection of model parameters and further corroborates the simulation results.

[0143] The above embodiments, in conjunction with the accompanying drawings, fully demonstrate the theoretical correctness of the method of the present invention, its effectiveness in simulation, and its superiority over traditional methods, possessing high engineering applicability and industrial application value.

[0144] It is hereby declared that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A modeling method for a unified cooperative driving model of heterogeneous vehicles in a tunnel entrance area, characterized in that, Includes the following steps: S1. Construct a dynamic model of human-vehicle driving considering tunnel environmental disturbances; S2. Design a dynamic weight allocation mechanism based on communication quality awareness; S3. Establish a driving model for connected autonomous vehicles in the tunnel entrance area by combining a dynamic weighting mechanism; S4. Introduce a communication degradation mechanism and establish an autonomous vehicle model; S5. Integrate human-driven vehicle dynamics model, connected autonomous vehicle driving model and autonomous vehicle model to construct a unified collaborative driving model for heterogeneous vehicles.

2. The modeling method for a unified cooperative driving model of heterogeneous vehicles in a tunnel entrance area according to claim 1, characterized in that, Step S1 includes the following sub-steps: S1.1 Calculate the self-driving force experienced by the driver during the driving process; In the formula, This represents the self-driving force experienced by the nth vehicle at time t, with the positive direction pointing from outside the tunnel to inside. This represents the mass of the nth vehicle; This represents the vehicle's desired maximum acceleration. This represents the speed of the nth vehicle at time t; δ represents the desired speed of the road; S1.2 is the adjustment index; S1.2 calculates the repulsive force between the driver and the vehicle. In the formula, This represents the repulsive force experienced by the nth vehicle at time t, with the positive direction pointing from outside the tunnel to inside. This represents the actual distance between the nth vehicle and the vehicle in front. This represents the expected safe distance between the nth vehicle and the vehicle in front. This is the expected headway. l is the position of the nth vehicle at time t; l is the vehicle length. γ is the minimum headway; b is the desired deceleration of the vehicle; γ is the adjustment index; It is the speed difference between the nth car and the (n-1)th car; S1.3 Calculate the potential force of the tunnel, i.e., the tunnel virtual force; In the formula, This represents the virtual force exerted on the nth vehicle by the driver at time t, with the positive direction pointing from outside the tunnel to inside. It is a weighted estimate of environmental disturbance based on speed limit gradient, rate of change of light intensity and linear index; σ is the spatial attenuation coefficient; d is the distance from the tunnel entrance, that is, the distance between the current position of the vehicle and the tunnel entrance, where d=0 at the entrance; μ is the distribution center parameter; This is the speed limit inside the tunnel; S1.4 Construct a dynamic model of human-vehicle driving in the tunnel entrance area, that is, characterize the dynamic behavior of human-vehicle driving in the tunnel entrance area by the self-driving force, the repulsive force between vehicles, and the tunnel virtual force during the driving process. The acceleration model of the nth vehicle in the longitudinal direction at the tunnel entrance area is as follows: In the formula, This represents the longitudinal acceleration of the nth vehicle driven by the driver.

3. The modeling method for a unified cooperative driving model of heterogeneous vehicles in a tunnel entrance area according to claim 2, characterized in that, The specific content of step S2 is as follows: For CAVs, a dynamic weight allocation mechanism for their interaction with preceding vehicle information is designed. This dynamic weight allocation mechanism dynamically calculates the fusion weight of each preceding vehicle's information based on real-time communication quality indicators, relative motion state, and distance information. The calculation formula is as follows: In the formula, This represents the normalized weight of the k-th vehicle at the current moment; k is the preceding vehicle number. The fusion coefficient has a range of values. This reflects the ratio of the current state to the historical weights; The basic weights of the k-th vehicle in the initial topology are based on the weights of the discretized signal-to-noise ratio (SNR) and packet loss rate (PLR) in the CPT table, reflecting the fundamental impact of communication quality on information reliability. This is the distance decay factor, reflecting the impact of the distance to the vehicle in front on the weight. For speed correction factor, It is the current speed of the vehicle. It is the speed of the kth car in front. The maximum speed limit for roads is used to reflect the impact of relative speed on communication quality; The number of valid preceding vehicles; This represents the normalized weight of the k-th vehicle at the previous moment.

4. The modeling method for a unified cooperative driving model of heterogeneous vehicles in a tunnel entrance area according to claim 3, characterized in that, Step S3 includes the following sub-steps: S3.1 Calculate the self-driving force experienced by the connected autonomous vehicle (CAV) during driving; In the formula, Let represent the self-driving force experienced by the i-th connected autonomous vehicle at time t, with the positive direction pointing from outside the tunnel to inside the tunnel; Indicate the mass of the i-th vehicle; This represents the speed of the i-th vehicle at time t; S3.2 Calculate the repulsive force in the connected autonomous driving workshop; In the formula, Let represent the repulsive force experienced by the i-th connected autonomous vehicle at time t, with the positive direction pointing from outside the tunnel to inside the tunnel; This represents the actual distance between the i-th vehicle and the vehicle in front; This represents the expected safe distance between the i-th vehicle and the vehicle in front; It is the position of the i-th vehicle at time t; It is the speed difference between the i-th car and the (i-1)-th car; S3.3 Calculation of multi-vehicle cooperative feedback terms; In the formula, For multiple preceding vehicle collaborative feedback items; The dynamic weight corresponding to the kth preceding vehicle is obtained through step S2; σ represents the total delay caused by information transmission and processing; σ is the adjustment parameter for the speed difference of the preceding vehicle. S3.3 Establish a driving model for connected autonomous vehicles in the tunnel entrance area; That is, the acceleration model of the i-th connected autonomous vehicle: In the formula, Let represent the acceleration of the i-th vehicle.

5. The modeling method for a unified cooperative driving model of heterogeneous vehicles in a tunnel entrance area according to claim 4, characterized in that, The specific content of step S4 is as follows: Define communication status indicator variables When the signal-to-noise ratio (SNR) of a connected autonomous vehicle (CAV) is lower than a threshold or the packet loss rate (PLR) is higher than a threshold, the CAV's communication function is deemed to be faulty, and the setting is... The vehicle degenerates into an autonomous vehicle (AV) that relies solely on onboard sensors; The AV employs a radar-based adaptive cruise control strategy, i.e., an autonomous vehicle model, whose control inputs are... for: in, , To control the gain; Desired vehicle spacing; This represents the desired time interval.

6. The modeling method for a unified cooperative driving model of heterogeneous vehicles in a tunnel entrance area according to claim 5, characterized in that, The specific content of step S5 is as follows: Integrate the human-driven vehicle dynamics model, the connected autonomous vehicle driving model, and the autonomous vehicle model established in steps S1-S4, and introduce vehicle type indicator variables. and communication status indicator variables A unified acceleration model is constructed, namely, a unified cooperative driving model for heterogeneous vehicles, with the following expression: In the formula, For vehicle type indicator variables, =1 indicates CAV / AV, =0 indicates HV; This is the self-driving force term, related to the vehicle's current speed. Related, propelling the vehicle forward; The term represents the repulsive force between vehicles, based on the distance between vehicles. Generates information to describe the interactions between vehicles.

Images