Inter-vehicle Communication Research Articles

Smart Collaborative Intrusion Detection System for Securing Vehicular Networks Using Ensemble Machine Learning Model

The advent of the Fourth Industrial Revolution has positioned the Internet of Things as a pivotal force in intelligent vehicles. With the source of vehicle-to-everything (V2X), Internet of Things (IoT) networks, and inter-vehicle communication, intelligent connected vehicles are at the forefront of this transformation, leading to complex vehicular networks that are crucial yet susceptible to cyber threats. The complexity and openness of these networks expose them to a plethora of cyber-attacks, from passive eavesdropping to active disruptions like Denial of Service and Sybil attacks. These not only compromise the safety and efficiency of vehicular networks but also pose a significant risk to the stability and resilience of the Internet of Vehicles. Addressing these vulnerabilities, this paper proposes a Dynamic Forest-Structured Ensemble Network (DFSENet) specifically tailored for the Internet of Vehicles (IoV). By leveraging data-balancing techniques and dimensionality reduction, the DFSENet model is designed to detect a wide range of cyber threats effectively. The proposed model demonstrates high efficacy, with an accuracy of 99.2% on the CICIDS dataset and 98% on the car-hacking dataset. The precision, recall, and f-measure metrics stand at 95.6%, 98.8%, and 96.9%, respectively, establishing the DFSENet model as a robust solution for securing the IoV against cyber-attacks.

Machine-Learning-Based Path Loss Prediction for Vehicle-to-Vehicle Communication in Highway Environments

Vehicle-to-vehicle (V2V) communication, which plays an important role in intelligent transportation systems, has been statistically proven to improve traffic efficiency and reduce the probability of accidents. In real-world applications, it is critical to accurately estimate the path loss parameter in communication channels due to the variable and complex propagation environments often encountered in inter-vehicle communication scenarios. This paper presents a study on various machine learning methods to improve path loss estimation in V2V communication using a dataset (192,000 observations) obtained from field measurements of highway environments in the Trabzon and Gümüşhane provinces in Türkiye. For this purpose, path loss estimation was carried out with different machine learning algorithms such as Artificial Neural Networks, Random Forest, Linear Regression, Gradient Boosting, Support Vector Regression, and AdaBoost by using various environmental and system features. Then, performance comparisons were conducted between machine learning methods and traditional empirical approaches such as log-distance, two-ray, and log-ray. Examining the outputs reveals that machine learning methods outperform traditional methods and yield results quickly. As a result, the Random Forest and Gradient Boosting methods demonstrated the highest prediction performances, with R2 values of 0.97 and 0.96, MAE values of 0.0557 and 0.0701, and RMSE values of 0.0774 and 0.0964, respectively, outperforming both empirical methods, other machine learning techniques, and the existing studies based on V2V. Overall, our study provides significant contributions to the existing literature by providing a comprehensive parameter set for highway environments, examining the path loss prediction performance of machine learning models with different capabilities, and comparing them with traditional methods. This study not only fills a critical gap in the existing literature but also highlights the necessity, efficiency, and originality of machine learning approaches for improving reliable V2V communication systems.

Adaptive event-triggered sliding mode control for platooning of heterogeneous vehicular systems and its [formula omitted] input-to-output string stability

Platooning of vehicular systems is an effective technique for enhancing transportation efficiency. As the scale of the vehicular platoon systems increases, disturbances on individual vehicles can affect the whole platoon through their connections. Besides, excessive vehicles impose a significant burden on communication devices. Towards this end, this work investigates the distributed platoon control problem of connected vehicular systems subject to disturbances by employing a resource-efficient communication mechanism. The proposed adaptive event-triggered mechanism (AETM) avoids periodic data transmission and reduces communication burden among vehicles. Besides, the AETM regulates the triggered threshold dynamically via the perception of spacing errors and avoids continuous inter-vehicle communication. Next, an AETM-based finite-time extended state observer (AFESO) is designed to alleviate the impact of the external disturbances. Then, an adaptive event-triggered distributed sliding mode control (DSMC) framework is developed to guarantee platoon stability. It is approved that, under the proposed control method, the closed-loop system subject to the disturbances satisfies the L2 input-to-output string stability (L2-IOSS). The salient feature of the AETM-based DSMC is that the AETM can effectively reduce communication consumption, while DSMC mitigates the performance degradation caused by triggering errors and disturbances. Finally, numerical simulations demonstrate the effectiveness of the proposed algorithm.

BeHarmony: Blockchain-Enabled Trustworthy Communication and Legitimate Decision Making in Multi-Party Internet of Vehicles Systems

The rapid development of the Internet of Vehicles using centralized systems faces significant challenges, including reliability and security vulnerabilities and high latency. This paper introduces a blockchain-enabled authentication and communication network for scalable IoV to enhance security, reduce latency, and relieve the dependency on centralized infrastructures. The network applies blockchain-enabled domain name services and mutual authentication for fault tolerance consensus, such as PBFT and RAFT, featuring a primary layer of road side units and edge servers for inter-vehicle communication and a sub-layer within each vehicle for intra-vehicle communication. The study evaluates various scenarios and assesses roadside unit availability based on random distribution along vehicle routes. This paper also discusses the legal issues involved in the proposed model, highlighting that the IoV system should be governed by a contract-based decentralized IoV system comprising both smart contracts and traditional contracts. This model offers a novel approach to developing a decentralized, secure, efficient, and ethical IoV ecosystem, advancing autonomous and reliable vehicular networks.

A deep reinforcement learning-based intelligent QoS optimization algorithm for efficient routing in vehicular networks

With the rapid development of Telematics, Vehicle Self-Organizing Networks (VANETs) play an increasingly critical role in Intelligent Transportation Systems (ITS). Especially in the environment without roadside assistance units (RSUs), how to effectively manage inter-vehicle communication and improve the stability and communication efficiency of the network has become a hot topic of current research. In this paper, a Deep Reinforcement Learning-based Intelligent QoS-optimized efficient routing algorithm for vehicular networks (DRLIQ) is proposed for VANETs with/without RSU environments, and routing methods are proposed respectively. Among them, in RSU-free environment, the DRLIQ algorithm utilizes the powerful processing capability of deep reinforcement learning to intelligently select the optimal data transmission path by dynamically learning and adapting to the changes in the vehicular network, thus effectively reducing communication interruptions and delays, and improving the accuracy of data transmission. The performance of the DRLIQ algorithm under different vehicle densities is evaluated in simulation experiments and compared with current popular algorithms. The experimental results show that the DRLIQ algorithm outperforms the comparison algorithms in reducing the number of communication interruptions, BER and network delay, especially in vehicle-dense environments. In addition, the DRLIQ algorithm shows higher adaptability and stability in coping with network topology changes and vehicle dynamics.

A Grid-based Framework for Managing Autonomous Vehicles' Movement at Intersections

With a full fleet of autonomous vehicles (AV) in the future, the concept of signal-free intersections has attracted transportation researchers. Multiple studies have investigated different methods and protocols to facilitate vehicle decision-making at intersections. Most of these methods have relied on strong centralized or inter-vehicle communications. This research aims to develop a new protocol for managing vehicle movement at four-leg intersections assuming a fully automated fleet. The main concept of the proposed protocol is to maintain a continuous movement of vehicles entering the intersection, without stopping in queues, by controlling their sequence of movements. In the methodology a dynamic occupancy grid (DOG) approach is applied by initially dividing the intersection into dynamic grids (i.e. cells). The cells are in a virtual movement, emanating away from each lane-group without overlapping, like a gear machine. Each vehicle sits on a specific cell to traverse the intersection safely at a predetermined time. In other words, vehicles approaching an intersection must register their speed and position by certain sensors, and in return receive the appropriate acceleration and speed to finally be allocated to the suitable moving grids. The efficiency of the applied protocol was demonstrated by a practical example that presented a higher intersection capacity value exceeding the ideal saturation flow rate in some cases. This reflects the efficiency of the implemented protocol and its applicability to different low to high traffic flows. Moreover, the protocol shows more flexibility in dealing with different weather, geometric and traffic conditions.

Artificial intelligence of things for smart cities: advanced solutions for enhancing transportation safety

In the context of smart cities, ensuring road safety is crucial due to increasing urbanization and the interconnected nature of contemporary urban environments. Leveraging innovative technologies is essential to mitigate risks and create safer communities. Thus, there is a compelling imperative to develop advanced solutions to enhance road safety within smart city frameworks. In this article, we introduce a comprehensive vehicle safety framework tailored specifically for smart cities in the realm of Artificial Intelligence of Things (AIoT). This framework seamlessly integrates a variety of sensors, including eye blink, ultrasonic, and alcohol sensors, to bolster road safety. The utilization of eye blink sensor serves to promptly detect potential hazards, alerting drivers through audible cues and thereby enhancing safety on smart city roads. Moreover, ultrasonic sensors provide real time information about surrounding vehicle speeds, thereby facilitating smoother traffic flow. To address concerns related to alcohol consumption and its potential impact on road safety, our framework incorporates a specialized sensor that effectively monitors the driver’s alcohol levels. In instances of high alcohol content, the system utilizes GPS and GSM technology to automatically adjust the vehicle’s speed while simultaneously notifying pertinent authorities for prompt intervention. Additionally, our proposed system optimizes inter-vehicle communication in smart cities by leveraging Li-Fi technology, enabling faster and more efficient data transmission via visible light communication (VLC). The integration of Li-Fi enhances connectivity among connected vehicles, contributing to a more cohesive and intelligent urban transportation network. Through the structured integration of AIoT technologies, our framework lays a robust foundation for a safer, smarter, and more sustainable future in smart city transportation. It offers significant advancements in road safety and establishes the groundwork for further enhancement in intelligent urban transportation networks.

Development of an Arduino-Integrated Wireless Inter-Vehicle Communication Infrastructure for Enhanced Traffic Safety and Efficiency

This project addresses the escalating challenges of road traffic management by proposing a sophisticated vehicle-to-vehicle (V2V) communication system based on Arduino technology. The growing congestion and safety concerns demand innovative solutions to facilitate real-time information exchange among vehicles. Our objective is to design and implement a robust V2V communication infrastructure utilizing Arduino microcontrollers and wireless modules. By establishing seamless communication channels between vehicles, we aim to enhance road safety, optimize traffic flow, and reduce accidents. This abstract outlines the problem statement of inadequate communication systems in current traffic management and presents our proposed solution to mitigate these issues through Arduino-enabled V2V communication.

A scalable, dynamic, and secure traffic management system for vehicular named data networking applications

Several intelligent transport systems in vehicular networks (VANETs) rely on IP-based models for inter-vehicle communication. However, many approaches for Cooperative Intelligent Transport Systems and distributed systems are not secure and appropriate for VANETs, using specific technologies, centralized network architectures, or limited communication models. In addition, many applications are forced to deal with complex services that the network layer does not provide. This paper proposes NDN-Waze, a comprehensive distributed and lightweight system for vehicular traffic management using Named Data Networking as the main underlying communication protocol. NDN-Waze is a proof-of-concept design that has applications in vehicular route guidance, traffic and mobility monitoring, as well as roadside units as a service in urban areas. We use the NDN-Waze route guidance module to compare the VANET model via Named Data Networking (NDN) and Internet Protocol (IP). Our study employs traditional network metrics to support the hypothesis that the communication architecture has a direct impact on both efficiency and the quality of service operations. The simulation results demonstrate that NDN properties can override traditional IP-based forwarding proposals in VANET.

Araç Araç Haberleşmesinde Blokzincir Kullanımı Üzerine Bir İnceleme

In this study, there are theoretical information about security in communication between block chain and vehicles. When talking about block chain, its types, features and content are mentioned. In the light of all this theoretical information, the studies made using block chain in inter-vehicle communication are examined. In addition, information about the methods used and their contents related to these studies are given. The theoretical background of this information is explained under general headings. In addition, the methods used by the studies examined are also categorized and explained together with the studies. In the study, first of all, detailed information is given about the block chain as stated. Afterwards, VANET (Vehicles Ad-hoc Network) security is mentioned. Afterwards, the studies in the literature are explained, and finally, future studies are mentioned in the conclusion section.

An Anti-Collision Algorithm for Self-Organizing Vehicular Ad-Hoc Network Using Deep Learning

The rapid increase of the number of motor vehicles over the last several decades has driven a corresponding increase in the severity of traffic congestion, which has exhibited a considerable human impact. Intelligent anti-collision control via inter-vehicle communication technology can facilitate vehicles adhering to spacing speed standards and improve road utilization and traffic efficiency. In this paper, we apply a deep learning image estimation model based on joint attention mechanism. The network framework uses a deep estimation network Yolov5 and a location based VANET information fast transmission strategy to work together. This paper mainly considers vehicle distance measurement technology as a point of entry to study multi-sensor information fusion for vehicle collision prevention technology based on the self-organizing vehicular ad-hoc network (VANET). This paper also proposes a solution based on the strategy of one-way transmission of shared information and dynamic valuation of a cluster distance threshold with vehicle density. The proposed vehicle anti-collision control algorithm is designed to realize dynamic vehicle control via inter-vehicle communication. In this paper, the two-way coupling of traffic flow and network simulator is used to randomly generate vehicle nodes on the road, and the behavior of the anti-collision system is simulated. The experimental results show that the predetermined control goal is achieved, which demonstrates the effectiveness of the proposed algorithm.

Cybersecurity for autonomous vehicles against malware attacks in smart-cities

Smart Autonomous Vehicles (AVSs) are networks of Cyber-Physical Systems (CPSs) in which they wirelessly communicate with other CPSs sub-systems (e.g., smart -vehicles and smart-devices) to efficiently and securely plan safe travel. Due to unreliable wireless communication among them, such vehicles are an easy target of malware attacks that may compromise vehicles’ autonomy, increase inter-vehicle communication latency, and drain vehicles’ power. Such compromises may result in traffic congestion, threaten the safety of passengers, and can result in financial loss. Therefore, real-time detection of such attacks is key to the safe smart transportation and Intelligent Transport Systems (ITSs). Current approaches either employ static analysis or dynamic analysis techniques to detect such attacks. However, these approaches may not detect malware in real-time because of zero-day attacks and huge computational resources. Therefore, we introduce a hybrid approach that combines the strength of both analyses to efficiently detect malware for the privacy of smart-cities.

BlockAuth: A blockchain-based framework for secure vehicle authentication and authorization.

Intelligent Transport System (ITS) offers inter-vehicle communication, safe driving, road condition updates, and intelligent traffic management. This research intends to propose a novel decentralized "BlockAuth" architecture for vehicles, authentication, and authorization, traveling across the border. It is required because the existing architects rely on a single Trusted Authority (TA) for issuing certifications, which can jeopardize privacy and system integrity. Similarly, the centralized TA, if failed, can cause the whole system to collapse. Furthermore, a unique "Proof of Authenticity and Integrity" process is proposed, redirecting drivers/vehicles to their home country for authentication, ensuring the security of their credentials. Implemented with Hyperledger Fabric, BlockAuth ensures secure vehicle authentication and authorization with minimal computational overhead, under 2%. Furthermore, it opens up global access, enforces the principles of separation of duty and least privilege, and reinforces resilience via decentralization and automation.

Decentralized Receiver-based Link Stability-aware Forwarding Scheme for NDN-based VANETs

The inter-vehicle communication in the vehicular ad-hoc network (VANET) providing ubiquitous mobility to mobile users is considered an essential component of intelligent transportation systems (ITS), enabling a wide range of safety and non-safety applications in which the vehicles seamlessly interact and co-operate with each other. The recent paradigm shift in VANETs protocols has led to the adoption of name data networking (NDN) as an underlying communication protocol in VANETs that replaces traditional address-based communication philosophy, treats content as a first-class citizen, and assigns unique names to contents consequently enabling name-based routing and forwarding, caching, and security. Although NDN has significantly improved the underlying communication issues in VANETs, a plethora of challenges such as message broadcast storms, in-efficient packet suppression mechanisms, intermittent connectivity owing to frequent topological changes, and reverse path partioning (RPP) require further attention from the research community. To cope with these issues, this paper proposes decentralized receiver-based link stability-aware forwarding (DRLSF) protocol. The DRLSF is a beacon-less receiver-based multi-hop protocol, well-suited for pull-based applications where end users request desired data by sending packets. The comparative performance analysis validates the effectiveness of the DRLSF protocol in different VANET environments and application scenarios.

Multi-agent reinforcement learning for safe lane changes by connected and autonomous vehicles: A survey

Connected Autonomous vehicles (CAVs) are expected to improve the safety and efficiency of traffic by automating driving tasks. Amongst those, lane changing is particularly challenging, as it requires the vehicle to be aware of its highly-dynamic surrounding environment, make decisions, and enact them within very short time windows. As CAVs need to optimise their actions based on a large set of data collected from the environment, Reinforcement Learning (RL) has been widely used to develop CAV motion controllers. These controllers learn to make efficient and safe lane changing decisions using on-board sensors and inter-vehicle communication. This paper, first presents four overlapping fields that are key to the future of safe self-driving cars: CAVs, motion control, RL, and safe control. It then defines the requirements for a safe CAV controller. These are used firstly to compare applications of Multi-Agent Reinforcement Learning (MARL) to CAV lane change controllers. The requirements are then used to evaluate state-of-the-art safety methods used for RL-based motion controllers. The final section summarises research gaps and possible opportunities for the future development of safe MARL-based CAV motion controllers. In particular, it highlights the requirement to design MARL controllers with continuous control for lane changing. Moreover, as RL algorithms by themselves do not guarantee the level of safety required for such safety-critical applications, it offers insights and challenges to integrate safe RL methods with MARL-based CAV motion controllers.

Advanced adaptive cruise control and vehicle to vehicle communication using LiDAR

The ACC system is a type of driving assistance that helps reduce the burden on a driver. It operates by controlling the vehicle's acceleration and deceleration to maintain a certain speed or avoid a collision. In order to improve the accuracy of its operation, this paper proposes a method that uses radar to detect the presence of the preceding vehicle. It also takes into account the vehicle sideslip angle and the distance between the two vehicles. The proposed method can help improve the driving stability. In addition, we are currently using the vehicle-to-vehicle communication technology in our system. This type of inter-vehicle communication is not widely used yet and is a new paradigm in the field of communication. One of the main advantages of this type of communication is that it doesn't rely on third-party networks such as cellular networks. Despite the various advantages of this technology, it still has a long way to go before it can be fully deployed. The increasing number of operators who are interested in this technology has led to the development of more effective and efficient V2V solutions. This paper aims to provide a comprehensive analysis of the various challenges and benefits of this type of communication. The paper also focuses on the various security issues that are associated with this technology. One of these is the potential interference. This issue can prevent the system from operating efficiently. In order to address this issue, the paper suggests an improved architectural solution that can help prevent this type of interference. The proposed work also proposes a standardized space for the various components of the V2V system. This will allow them to easily integrate into the system. Vehicle-to-Vehicle communication is a type of wireless communication that is designed to allow cars to talk to one another.

A comprehensive and configurable simulation environment for supporting vehicular named-data networking applications

The Named Data Networking (NDN) architecture, with its network-layer features, services, and properties, is well-suited for vehicular applications and inter-vehicle communication (IVC). In contrast, IP-based host-centric architectures struggle with challenges inherent to the vehicular ad-hoc network (VANET) context, such as node mobility, data security, efficient data forwarding, and routing. Named Data Networking takes a fundamental departure from today’s IP-based architectures for VANET/IVC and thus requires extensive experimentation and evaluation. To facilitate experimentation with Vehicular Named-Data Networking, we present the NDN4IVC, an open-source simulator/framework that facilitates testing of more realistic VANET applications via the NDN stack. This project utilizes two popular simulators for VANET simulation: the NS3 network simulator with the ndnSIM module and SUMO, a simulator of urban mobility. NDN4IVC allows real-time bidirectional communication between SUMO and NS3 to support more data about road traffic and vehicular mobility. The framework can model the impact of vehicular networks on road traffic and investigate complex interactions between the two domains. We included two sample applications in VANET to demonstrate how different NDN properties can be used. Experiments were conducted as proof-of-concept studies to demonstrate the potential of the framework and its functionality.

On the assessment of the dynamic platoon and information flow topology on mixed traffic flow under connected environment

It has been well documented that the mixed traffic flow of human-driven vehicles (HDVs) and connected and automated vehicles (CAVs) causes an unsound communication environment and weakens the beneficial effect of CAVs. However, the communication heterogeneity in the mixed traffic, which leads to incomplete information flow topology (IFT) and further affects the processes of vehicle control and platoon operation, did not receive its deserved attention. In contrast to previous research related to the IFT and platoons in static size, this study introduced the multi-anticipation and dynamic platoons into a more complex mixture of CAVs and HDVs and jointly considered the mode transition and driver uncertainties, e.g., estimation error and driving compliance. Specifically, we proposed multi-class generic controls to represent the basic longitudinal maneuver for each type of vehicle and developed customized car-following dynamics for each controller. In order to better adapt to the communication heterogeneity in the mixed traffic, we defined a novel controller for CAVs, namely mixed AV-CAV vehicle (MACV) controller and developed the corresponding control algorithm, namely AV-CAV integrated control (ACIC) with multi-source information from both sensors system and inter-vehicle communication. Besides, another important point worth considering in mixed traffic is the spatial distribution of CAVs in platoons. As such, three platoon policies, i.e., the cross policy, the continuum policy, and the random policy, were investigated. To our best knowledge, this is a new attempt to discuss the impact of platoon policies on mixed traffic, with a combination of different IFTs. Finally, multiple traffic flow characteristics, covering robustness, efficiency, safety, and emission, were assessed by calculating their corresponding surrogate measures. Results showed that the proposal of MACV controller and ACIC was expected to efficiently prevent the deleterious degradation from the CAV controller to AV controller and construct a more stable transition for changes in vehicle controls. Furthermore, a high proportion of communicable vehicles could significantly reduce the heterogeneity of mixed traffic and improve traffic flow characteristics. Besides, this study demonstrated the outstanding performance of multi-anticipative IFTs compared to the single-anticipative IFT and revealed the advantage of cross policy in traffic robustness and safety. These findings can help provide feasible solutions for IFTs, platoon operation, and vehicle control in the mixed traffic flow and identify their respective benefits in terms of traffic flow characteristics.

Autonomous Data Collection With Dynamic Goals and Communication Constraints for Marine Vehicles

In marine robotics, data-collection operations often require an autonomous underwater vehicle (AUV) to collaborate with an unmanned surface vehicle (USV). The mission for the AUV is to reach many goal locations, avoid obstacles and unsafe areas, and maintain communication with the USV. The goals, however, are not known a priori, but are dynamically discovered by the USV as it moves along a predefined path. The USV communicates the discovered goals to the AUV along with rewards for reaching each goal to incentivize the AUV to increase the sum of the rewards when obstacles, time, and communication constraints make it impossible to reach all the goals. We develop a framework comprised of an execution module and a multi-layered planner to enable the AUV to avoid collisions, maintain communication with the USV, and increase the sum of the rewards by reaching many of the discovered goals. The execution module enables the AUV to follow the planned motions, invoking the planner when new goals are discovered. To facilitate navigation, the planner constructs a 3D roadmap that captures the connectivity of the environment by sampling and connecting waypoints in the free space. The high-level planning layer is based on informed discrete search to find roadmap paths that satisfy the communication constraints and increase the sum of the goal rewards. The low-level layer uses sampling-based motion planning to expand a tree of feasible motions along these roadmap paths. The layers interact to update the planned motions as new goals are discovered. Experiments using 3D environments and an increasing number of goals demonstrate the efficiency of the approach to solve dynamic multi-goal motion-planning problems with communication constraints. <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Note to Practitioners</i> —This paper is motivated by the problem of at-sea data collection using unmanned vehicles where inter-vehicle communications constraints must be maintained. This is a challenging problem that has generally required significant human monitoring and intervention due to the communication constraints and planning challenges. In this paper, we leverage recent advances in underwater communications and focus on new problems that arise in the planning space, specifically, how we generate trajectories for an AUV that satisfies communication range constraints while still performing an underlying task. We show that it is possible to plan for an AUV in real-time while considering these constraints. This paper suggests that our approach can support these complex at-sea missions. Future research will focus on field deployments and introducing more uncertainty into the underlying sensor models during planning.

Reinforcement Learning-Based Approach for Minimizing Energy Loss of Driving Platoon Decisions.

Reinforcement learning (RL) methods for energy saving and greening have recently appeared in the field of autonomous driving. In inter-vehicle communication (IVC), a feasible and increasingly popular research direction of RL is to obtain the optimal action decision of agents in a special environment. This paper presents the application of reinforcement learning in the vehicle communication simulation framework (Veins). In this research, we explore the application of reinforcement learning algorithms in a green cooperative adaptive cruise control (CACC) platoon. Our aim is to train member vehicles to react appropriately in the event of a severe collision involving the leading vehicle. We seek to reduce collision damage and optimize energy consumption by encouraging behavior that conforms to the platoon's environmentally friendly aim. Our study provides insight into the potential benefits of using reinforcement learning algorithms to improve the safety and efficiency of CACC platoons while promoting sustainable transportation. The policy gradient algorithm used in this paper has good convergence in the calculation of the minimum energy consumption problem and the optimal solution of vehicle behavior. In terms of energy consumption metrics, the policy gradient algorithm is used first in the IVC field for training the proposed platoon problem. It is a feasible training decision-planning algorithm for solving the minimization of energy consumption caused by decision making in platoon avoidance behavior.