A communication and cooperative control method for a connected automatic driving vehicle in a mixed traffic flow

Abstract

The application discloses a communication and cooperative control method for networked automatic driving vehicles for mixed traffic flow, which is applied to the field of intelligent vehicle networking and aims at solving the problem that the existing technology is difficult to balance the limitation of communication resources and the information updating demand of digital twinning and realize effective cooperation among CAVs; the application constructs a distributed digital twinning system in each CAV, monitors and dynamically simulates the driving strategy of the road condition, the communication channel state and the surrounding vehicles in real time, makes up for the decision deficiency caused by the single-vehicle perception limitation, and thus realizes the group intelligence cooperation among vehicles. While the individual driving style is reserved, the CAV can more effectively relieve the disturbance of HDV driving to the traffic, and significantly improve the efficiency and stability of the traffic flow. In addition, the application fully considers the pressure of the perception information sharing on the communication system in the cooperative driving process of the CAV, optimizes the information with the most decision value under the condition of limited communication resources, and dynamically adjusts the interaction frequency among the vehicles according to the channel state, so that the light-weight design is realized, and the applicability and reliability of the scheme in the actual application scene are further enhanced.

Description

Technical Field

[0001] This invention belongs to the field of intelligent vehicle networking, and specifically relates to a communication and cooperative control technology for autonomous vehicles. Background Technology

[0002] Vehicle-to-everything (V2X) technology and high-performance onboard computing platforms enable connected autonomous vehicles (CAVs) to communicate in real time and process driving information efficiently, allowing them to flexibly cope with complex and ever-changing road environments. This capability not only improves driving safety but also provides strong technical support for collaborative control between vehicles, potentially further enhancing traffic flow stability and improving road efficiency. Therefore, connected autonomous driving technology is widely regarded as one of the key driving forces propelling modern transportation towards intelligence and efficiency. However, despite significant progress in autonomous driving technology, its widespread adoption is still limited by factors such as economic costs, legal regulations, and social acceptance. In the current traffic environment, autonomous vehicles still need to coexist with human-driven vehicles (HDVs), forming a hybrid autonomous driving transportation system.

[0003] In such systems, human drivers' limited perception, delayed reaction times, and inaccurate decision-making often disrupt traffic flow. If these disruptions are not effectively controlled, they accumulate and eventually lead to traffic congestion. Furthermore, subjective factors such as individual driving habits and psychological states increase the unpredictability of human driver behavior, further complicating the management of hybrid automated driving traffic systems. Therefore, how to reduce HDV disturbances to traffic through collaborative control of CAVs (Continuously Operated Vehicles) while ensuring traffic safety, thereby improving road stability and traffic efficiency, has become a crucial issue that urgently needs to be addressed in the field of autonomous driving.

[0004] Existing research indicates that cooperative control of CAVs can improve traffic performance to some extent. However, most of these studies are based on idealized assumptions of complete information sharing and transparent driving strategies among vehicles, neglecting the finite nature of communication resources and the instability of physical channels. For example, in a typical intersection scenario with 20 vehicles, the communication bandwidth requirement for sharing radar data and camera images among vehicles can reach up to 1Gbps. Furthermore, the amount of information increases exponentially with the number of vehicles, which not only increases communication latency and packet loss rate but may also adversely affect driving performance. Therefore, in the cooperative control of connected autonomous driving, fully considering the limitations of communication resources is crucial for improving the practical feasibility and effectiveness of the strategy.

[0005] Digital twin technology offers a new solution for bridging the gap between communication and transportation systems. As a virtual mapping of the physical world, digital twins can collect data in real time, perform in-depth simulation analysis, and make intelligent decisions, demonstrating unique advantages in solving joint design problems of complex systems. However, applying digital twin technology in hybrid autonomous driving transportation systems also faces many challenges. On the one hand, the accurate modeling and decision-making of digital twins rely on the comprehensiveness, real-time nature, and accuracy of information, which contradicts limited communication resources. On the other hand, due to the personalization of driving behavior, each vehicle needs to maintain its own digital twin, and the trend of decentralization exacerbates the complexity of cooperation between vehicles, potentially leading to inconsistent driving decisions. Therefore, how to balance the limitations of communication resources with the information update needs of digital twins and achieve effective collaboration between CAVs is a current technical challenge in hybrid autonomous driving transportation systems. Summary of the Invention

[0006] To address the aforementioned technical problems, this invention proposes a communication and cooperative control method for connected autonomous vehicles oriented towards mixed traffic flow. Taking full account of communication resource constraints, this invention reduces the disturbance to traffic caused by human drivers through efficient cooperative perception and control among CAVs, ensuring travel safety while improving traffic flow stability and road capacity.

[0007] The technical solution adopted in this invention is: a communication and cooperative control method for connected autonomous vehicles oriented towards mixed traffic flow. The application scenario is: in a road segment containing closely following HDVs and CAVs, the CAVs have the ability to perceive surrounding road conditions and communication, and manage and allocate communication resources through roadside base stations to share perceived information; the specific implementation process includes the following steps:

[0008] S1. Based on the overall road condition representation within the communication range, the CAV's assessment of the channel state obtained through spectrum sensing, and the CAV's fitting of the driving strategies of surrounding vehicles, a distributed digital twin is established on the CAV's onboard computing platform.

[0009] The overall road condition representation within the communication range includes the current CAV's perception information and the perception information of other CAVs obtained through vehicle-to-vehicle communication; the perception information of each CAV includes the vehicle status of the CAV and the point cloud data of the surrounding driving environment perceived by the CAV.

[0010] Vehicle status includes the speeds of the vehicles in front and behind, and the distances between the CAV and the vehicles in front and behind;

[0011] S2. Each CAV senses the surrounding road conditions through its onboard radar and generates point cloud data of the surrounding driving environment.

[0012] S3. A multi-agent reinforcement learning method based on attention mechanism to obtain the communication and driving decisions of each CAV; specifically:

[0013] S31. The digital twin of each CAV includes: an online network, which includes an attention network, a policy network, and a value network;

[0014] S32. Take the overall road condition representation within the communication range of each CAV as the input feature in the attention network, and linearly map the input feature into query vector and key vector to obtain the attention weight of the current CAV to other CAVs.

[0015] S33. Calculate the update interval index for each CAV based on attention weight and communication resource availability.

[0016] S34. Based on the update interval, communication requirements are generated. The roadside base station allocates time-frequency resource blocks for communication between CAVs according to the communication requirements. Each CAV communicates according to the allocated video resource blocks, and a reward function value reflecting communication performance is obtained. And update the overall road condition representation in the digital twin;

[0017] S35. Using the updated overall road condition representation as input features in the attention network, calculate the hidden state. ;

[0018] S36, will Input policy network to obtain driving acceleration ;

[0019] S37, CAV is obtained according to step S36 Drive and calculate reward values ​​that reflect traffic performance. ;

[0020] S38, Incorporate vehicle status and attention weighting. , , Add to the experience replay pool;

[0021] S39. Repeat steps S2-S38, randomly sample the experience in the replay pool, calculate the loss function and backpropagate, and update the attention network, policy network and value network.

[0022] The beneficial effects of this invention are as follows: Compared with existing technologies, this invention constructs a distributed digital twin system within each CAV (Carrier Available Vehicle) to monitor and dynamically simulate road conditions, communication channel status, and the driving strategies of surrounding vehicles in real time. This compensates for the decision-making deficiencies caused by the limitations of single-vehicle perception, thereby achieving collective intelligent collaboration among vehicles. While preserving individual driving styles, CAVs can more effectively mitigate traffic disturbances caused by HDV (High-Depth Vehicle) driving, significantly improving traffic flow efficiency and stability. Furthermore, this invention fully considers the pressure on the communication system caused by sharing perception information during CAV cooperative driving. By prioritizing the most valuable information for decision-making under limited communication resources and dynamically adjusting the interaction frequency between vehicles according to channel status, a lightweight design is achieved, further enhancing the applicability and reliability of the solution in practical application scenarios. Attached Figure Description

[0023] Figure 1 This is the application scenario of the present invention.

[0024] Figure 2 This is a flowchart of a hybrid autonomous driving communication and cooperative control algorithm based on lightweight digital twins.

[0025] Figure 3 This is a schematic diagram of a multi-agent reinforcement learning model based on an attention mechanism. Detailed Implementation

[0026] To facilitate understanding of the technical content of this invention by those skilled in the art, the following description, in conjunction with the accompanying drawings, further illustrates the invention.

[0027] To address the disturbances to overall traffic flow caused by human drivers in hybrid autonomous driving traffic flows, and the problem of limited communication resources affecting the interaction of perceptual information required for cooperative driving in practical applications, this invention proposes a hybrid autonomous driving traffic control method based on lightweight digital twins. Each CAV is equipped with an independent digital twin system to compensate for the limitations of single-vehicle perception and simulate the driving behavior of surrounding vehicles. Given limited communication resources, the digital twin system selects the most critical perceptual information for vehicle decision-making and traffic control, interacting with the system through asynchronous updates to provide decision support for the vehicle, thereby effectively improving the stability and efficiency of traffic flow in hybrid driving scenarios. Figure 1 As shown, the application scenario of this invention is a road containing closely following CAVs and HDVs.

[0028] The technical solution of this invention comprises two parts: a lightweight digital twin architecture and a multi-agent reinforcement learning model based on an attention mechanism. Starting from the car-following behavior among vehicles and its cumulative traffic effects, and considering the individual-interest-oriented nature of vehicles during driving, this invention establishes a digital twin system on each vehicle's onboard computing platform. This system virtually maps road conditions, channel conditions, and the driving strategies of surrounding vehicles, enabling vehicles to grasp more comprehensive road condition information, infer the driving behavior of other vehicles, and predict future road and channel conditions. This invention employs a multi-agent reinforcement learning model based on an attention mechanism. Based on the availability of current communication resources, vehicles distribute and select information interaction sources and frequencies, conduct real information interaction in physical channels, and use the interaction results to determine acceleration and deceleration for their driving.

[0029] The implementation process of this invention is as follows:

[0030] 1. Lightweight Digital Twin Architecture

[0031] In this invention, the set of all vehicles is represented as ,in Represents a set of HDVs. This represents the set of all CAV vehicles. For any following vehicle in a car-following pair... And the car in front If the following vehicle is an HDV, that is This is limited by the limited perception range of human drivers, and the following vehicle is always... The observation information is ,in The speed of the car in front. The distance between the two vehicles. Based on this perceived information, the following vehicle adjusts its speed to ensure a close following while maintaining safe driving conditions. Must meet:

[0032] (1);

[0033] in, and These represent the maximum deceleration of the front and rear vehicles during emergency braking, respectively. This refers to the time from when the driver of the following vehicle notices the vehicle in front braking to when they react by braking; during this time, the following vehicle maintains its original speed. Because HDV (High-Density Vehicle) driving decisions are based solely on the current state of the vehicle in front, it struggles to anticipate traffic conditions, resulting in poor performance in traffic flow stability and throughput. CAV (Conductor-Action Vehicle), on the other hand, can extend its perception range through onboard cameras and radar. If... Then its observation information is ,in and For the car behind The speed and distance between the two vehicles are dynamically measured by cameras, and This refers to the point cloud data of the surrounding driving environment generated by radar perception. In addition, CAVs can also exchange their perception information through vehicle-to-vehicle communication, giving the vehicle a wider field of vision.

[0034] However, frequent interactions place a significant burden on communication systems. Therefore, this invention establishes an independent digital twin system for each vehicle to achieve lightweight updating and maintenance of perception information. This allows each vehicle to efficiently acquire perception information from other vehicles with limited communication resources, thereby improving the efficiency of cooperative driving control. The construction process of the lightweight digital twin proposed in this invention mainly includes the following stages:

[0035] The first step is information acquisition. Information sources include vehicle-mounted cameras and radar sensing, surrounding vehicle perception acquired through vehicle-to-vehicle communication, assessments of communication resource availability during interactive sensing, and fitting of driving strategies of other vehicles during driving. Based on the acquired information, each CAV builds a distributed digital twin system on its own onboard computing platform. ,in This represents the vehicle's own perception, as well as the set of perception information obtained from other vehicles through vehicle-to-vehicle communication, used to form... A comprehensive representation of traffic conditions; This refers to the assessment of channel state obtained by the vehicle through spectrum sensing, which is used to support the vehicle's dynamic adjustment of information update requirements. yes Fitting the driving strategies of surrounding vehicles is used to infer the driving behavior of other vehicles, thereby improving the effectiveness of collaboration.

[0036] The second phase is the simulation phase. The digital twin architecture proposed in this invention differs from traditional digital twin systems, requiring real-time and accurate information maintenance and updates to support decision-making. This invention believes that not all data from other vehicles is equally important. Therefore, updates in the system are selectively synchronized to consider the correlation of channel states and the importance of exchanged information, thereby reducing the communication burden and building a more efficient and lightweight system. In the simulation phase, this invention focuses on how observation data from different vehicles affects the cooperative behavior of a single CAV. To more clearly represent the information exchange process, this invention uses update intervals. This is used to measure the time elapsed since the data was generated. Therefore, time is... Below It can be represented as By continuously interacting with and receiving feedback from the physical world, this invention analyzes the potential relationship between traffic utility and update intervals, thereby identifying the optimal information exchange that balances cooperative driving efficiency and communication constraints.

[0037] Finally, there is the decision-making and feedback phase. Based on the inferred update interval... Observational data is acquired from selected vehicles for continuous updating. Based on the updated , Distributed acceleration decisions are made, acting on the physical world and collecting feedback on overall traffic performance. This process not only supports real-time following decisions but also guides long-term driving strategies. As new information is continuously collected, vehicles respond by adjusting their interaction and driving strategies within their digital twins, enabling them to dynamically adapt to changing traffic conditions and communication resource constraints.

[0038] 2. Multi-agent reinforcement learning methods based on attention mechanisms

[0039] This invention proposes a multi-agent reinforcement learning method based on an attention mechanism to optimize the interaction of perceptual information between vehicles under conditions of limited communication resources, and to make driving decisions based on the interaction information. This method aims to improve road stability and traffic capacity while ensuring driving safety.

[0040] like Figure 3 As shown, each CAV is modeled as an agent connected by a wireless channel, and its communication and driving are defined as partially observable Markov processes. At each time step... Each vehicle first updates its local observation information. The representation of overall road conditions in the digital twin is then updated via vehicle-to-vehicle communication. Based on the information in the updated twin system, This further generates driving decisions, determining acceleration and deceleration behavior. Therefore, the core of this problem can be divided into two parts: the goal and frequency of information updates. Optimization, and decisions regarding driving acceleration generate.

[0041] In the attention mechanism, the input features of each vehicle These are linearly mapped to query vectors and key vectors for extracting relevant features. Vehicles For vehicles Attention weights are calculated as follows:

[0042] (2);

[0043] in, It is a query vector. It is a key vector. These are the dimensions of these vectors. Attention weights. The vehicle was identified. and vehicles The importance of information exchange between them, and The sum of attention weights for all vehicles equals 1.

[0044] While attention mechanisms can effectively assess the importance of observations from different CAVs, their assessment process relies on receiving data from other vehicles first, thus failing to fundamentally reduce redundant communication interactions. To address this issue, this invention fully utilizes the characteristics of Markov processes, incorporating the attention score from the previous time step as part of the current state. Compared to directly using attention weights, this method employs their reciprocal and rounded down to determine the frequency of information updates for each vehicle. The specific expression is as follows:

[0045] (3);

[0046] in, This represents the maximum number of vehicles each vehicle can communicate with at any given time, taking into account latency and bandwidth limitations. The formula is as follows:

[0047] (4);

[0048] in, This represents the system's total bandwidth limit. Indicates the size of the perceived data. The maximum permissible delay for successful transmission of perceived information. for and The expected value of the inter-channel gain, where For reference distance, The physical distance between the two vehicles. It is the path loss index. A Gaussian variable following a Rayleigh distribution represents the channel gain. To minimize average communication delay, orthogonal sub-channels are allocated between each vehicle by the roadside base station based on the communication requirements of each link, thereby reducing co-channel interference. This method rationally allocates the information update interval index for each vehicle by balancing bandwidth, delay, and physical communication conditions between vehicles, thus minimizing redundant communication interactions while ensuring communication efficiency.

[0049] counter Recorded from vehicle The duration since the last communication with other vehicles. When, it means Need and Communication to update its digital twin system The relevant parts. Subsequently, each CAV updates its [relevant information] via sidechain based on the communication resources allocated by the base station. And calculate the hidden state. :

[0050] (5);

[0051] `concatenate` concatenates multiple vectors, and `tanh` is the hyperbolic tangent function used for data activation. The entire calculation... The process is denoted as attention network , These are the parameters of the network. (In obtaining...) Afterwards, each vehicle can make independent decisions. Vehicle reward function. It can be divided into two parts from the perspective of transportation and communication, namely ,in Reflecting traffic performance, defined as:

[0052] (6);

[0053] ,express The absolute value of the change in following distance is used to measure the stability of traffic flow. for In time The driving speed is used to measure the road's capacity. This is to determine whether the following distance meets the minimum safe distance, which is the foundation of driving safety.

[0054] Reward function for communication performance Defined as:

[0055] (7);

[0056] This function is used to evaluate the communication latency of the vehicle when interacting with perceived information. Does it meet the maximum latency limit for successful transmission? This allows us to measure the efficiency of the communication system.

[0057] Due to acceleration It is a continuous variable. This invention designs an online attention network, a policy network, and a value network in the digital twin system of each CAV. (Policy network) Used for decision acceleration Value Network This is used to evaluate the quality of the actions output by the policy network. Furthermore, to enhance the stability of the policy, target network replicas are created for each of the three networks. , and It also synchronizes the parameters of the online network periodically through a soft update mechanism.

[0058] The loss function of the value network is:

[0059] (8);

[0060] (9);

[0061] in This is the decay factor. The loss functions for the attention network and the actor network can be calculated using the following formula:

[0062] (10);

[0063] Using the loss function described above, gradients can be calculated using the chain rule to update the relevant network parameters. This training process optimizes the model to derive information exchange and driving strategies that conform to communication resource constraints.

[0064] 3. Hybrid Automated Driving Traffic Control Algorithm Flow Based on Lightweight Digital Twin

[0065] This invention provides a cooperative control method for CAVs in hybrid autonomous driving scenarios. It fully considers the communication overhead of information sharing during cooperative driving, adaptively selects the objects and frequency of perception information sharing between vehicles, and implements cooperative control based on this information. The algorithm steps are as follows:

[0066] Step 1: Initialize road driving status, vehicle following relationships, communication resources, V2V links, etc.

[0067] Step 2: Build a lightweight digital twin system on the onboard computing platform of each CAV. Simultaneously initialize the online attention network, value network, and policy network in the twin system. , and the corresponding target network , and Initialize the counter ;

[0068] Step 3: CAV updates its perception using cameras and radar. ;

[0069] Step 4: Vehicle based on attention score ,calculate , as an indicator of the update interval;

[0070] Step 5: Determine which vehicles need to be retrieved for interaction information at the current moment by updating the interval and updating the counter;

[0071] Step 6: Based on communication requirements, the roadside base station allocates orthogonal sub-channels to the V2V links that need communication. Vehicles communicate according to the allocated communication resources and obtain communication-related reward values. And update the representation of road conditions in the digital twin. ;

[0072] Step 7, Based on the updated Calculating the hidden state for decision-making in a digital twin system ;

[0073] Step 8, Input Online Policy Network Gain driving acceleration ;

[0074] Step 9: The vehicle drives according to the decision-making driving behavior in the digital twin system, and calculates traffic-related reward values. ;

[0075] Step 10: Store the experience in the experience revisit pool, randomly sample, calculate the loss function, and update the online attention network, policy network, and value network;

[0076] Step 11: At regular intervals, use the parameters of the online network to soft update the target network;

[0077] Step 12, Return to Step 3;

[0078] Compared with other CAV cooperative control algorithms, this invention has the following innovations: First, for scenarios where CAVs and HDVs coexist, this invention establishes the relationship between CAV perception information interaction and cooperative control effects, proposes a lightweight digital twin architecture, and designs an information interaction strategy oriented towards traffic performance. Based on real-time traffic conditions and communication resource availability, the CAV adaptively selects the objects and frequencies of perception information interaction required for control through digital twin simulation, providing an optimized information transmission scheme for cooperative control under limited communication resources. Second, this invention proposes a multi-agent reinforcement learning method based on an attention mechanism for distributed cooperative control of CAVs. This method, while preserving the individual driving style and interests of CAVs, achieves multi-vehicle cooperation through swarm intelligence, effectively mitigating the disturbances to traffic flow caused by human driving behavior and improving traffic flow stability and efficiency.

[0079] Those skilled in the art will recognize that the embodiments described herein are intended to help the reader understand the principles of the invention, and should be understood that the scope of protection of the invention is not limited to such specific statements and embodiments. Various modifications and variations can be made to the invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the invention should be included within the scope of the claims of the invention.

Claims

1.A method for communication and cooperative control of connected automated vehicles in mixed traffic flow, characterized in that, The application scenario is: in a road section containing closely following HDVs and CAVs, the CAV has the ability to perceive the surrounding road conditions and communicate, and manages and allocates communication resources through roadside base stations to share perception information; the specific implementation process includes the following steps: S1, based on the overall road condition representation in the communication range, the evaluation of the channel state obtained by the CAV through spectrum sensing and the fitting of the driving strategy of the surrounding vehicles, a distributed digital twin is established on the vehicle-mounted computing platform of the CAV; The overall road condition representation in the communication range includes the perception information of the current CAV, the perception information of other CAVs obtained through vehicle-to-vehicle communication; the perception information of each CAV includes the vehicle state of the CAV and the surrounding driving environment point cloud data perceived by the CAV; The vehicle state includes the speed of the front and rear vehicles, and the distance between the CAV and the front and rear vehicles; S2, each CAV perceives the surrounding road conditions through a vehicle-mounted radar to generate surrounding driving environment point cloud data; S3, based on the attention mechanism multi-agent reinforcement learning method, the communication decision and driving decision of each CAV are obtained; specifically: S31, the digital twin of each CAV includes an online network, and the online network includes an attention network, a policy network and a value network; S32, the overall road condition representation in the communication range of each CAV is taken as the input feature in the attention network, the input feature is linearly mapped into a query vector and a key vector, and the attention weight of the current CAV to other CAVs is obtained; S33, according to the attention weight and the availability of communication resources, the index of the update interval of each CAV is calculated; the availability of communication resources is specifically: considering the delay and bandwidth limitation, the maximum number of vehicles that the current CAV can communicate at any time; the calculation formula of the index of the update interval is: ； wherein, denotes the update interval between the th CAV and the th CAV, denotes the maximum number of vehicles that can communicate at any time per vehicle, taking into account latency and bandwidth limitations, denotes the attention weight of the th CAV to the th CAV; S34, generating a communication demand based on the updated interval index, the wayside base station allocating time-frequency resource blocks for communication between the CAVs according to the communication demand, each CAV communicating according to the allocated time-frequency resource blocks, and obtaining a reward function value reflecting the communication performance and updating the overall road condition representation in the digital twin; S35, taking the updated overall road condition representation as input features in the attention network, calculating the hidden state ; S36, will Input policy network to obtain driving acceleration ; S37、CAV according to step S36 obtains driving and calculating a reward value reflecting traffic performance ; S38, putting the vehicle state, attention weight, , , into an experience replay pool; S39, repeat steps S2-S38, randomly sample the experience in the replay pool, calculate the loss function and back propagation, update the attention network, the policy network and the value network; The loss function specifically includes: The loss function of the value network is: ； ； wherein is a decay factor, is a reward function; a reward function of a vehicle includes two parts, i.e., a reward function of traffic performance and a reward function of communication performance, and is defined as wherein reflects traffic performance and is defined as ； representing the absolute value of the change in the following distance, for measuring the stability of the traffic flow; for the driving speed at the time , for measuring the traffic capacity of the road, is a judgment whether the following distance meets the minimum safety distance; Reward function for communication performance is defined as: ； Assessing communication latency for vehicles interacting with perception information Whether a latency upper bound for successful transmission is met thereby gauging efficiency of the communication system; The loss functions of the attention network and the policy network are calculated by the following formula: 。 2.The mixed traffic flow oriented communication and cooperative control method for connected and automated vehicles according to claim 1, wherein, The calculation formula of the availability of communication resources in step S3 is: ； wherein, is the total bandwidth upper limit of the system, denotes the size of the perception data amount, is the maximum allowed latency for successful transmission of the perception information, is the expected value of the channel gain between the th CAV and the th CAV, is the reference distance, is the physical distance between the th CAV and the th CAV, is the path loss exponent, denotes the channel gain, denotes the set of CAVs. 3.The hybrid traffic flow oriented connected automatic driving vehicle communication and cooperative control method of claim 2, wherein, Step S34 is specifically: the first CAV is denoted , a counter records the duration since the last communication with other CAVs, denotes the duration since the last communication with the first CAV, ; when , it means that the first CAV needs to be communicated with to update the overall road condition representation in its digital twin.

Images