Blockchain-based reputation model for vehicle platooning with common global goal

Abstract

Vehicle platooning is an active area of research in which vehicles travel closely together to improve fuel efficiency and safety, and relies heavily on the trustworthiness and cooperation of the participating vehicles. Traditional methods of managing trust in such systems face significant challenges, including vulnerability to malicious behavior and lack of transparency. Any such attack on the platoon from outside or inside will be disastrous as human lives are involved. Such attacks can be dealt with if we only allow trust worthy vehicles. In this research we propose a blockchain-based reputation model for vehicle platooning, designed to address these challenges and enhance the overall reliability and efficiency of platooning systems. Vehicles will only be allowed to join the platoon after calculating their reputation scores and these scores will be saved in a smart contract. This will ensure that every agent can reliably compute the score and no entity can compromise the working of the platoon.

Keywords

vehicle platooning smart contract blockchain multi-agent systems information security

1 Introduction

A Multi-agent system (MAS) consists of a collection of autonomous entities known as agents that work collaboratively within a shared environment to tackle complex problems that require coordination and distributed decision-making.¹ A MAS in vehicle platooning is an organized network of autonomous vehicles that function as individual agents. These agents collaborate through real-time communication, enabling distributed decision-making to achieve coordinated platooning. This system enhances fuel efficiency, safety, and traffic stability by synchronizing each vehicle's behavior while retaining a degree of autonomy.² The rapid advancement of autonomous and connected vehicle technologies has led to the concept of vehicle platooning, a coordination system where a group of vehicles follows a lead vehicle at a close, uniform distance. This approach enhances fuel efficiency, reduces emissions, and increases traffic safety by minimizing aerodynamic drag and optimizing road space.³ Platooning is a promising application in intelligent transportation systems, potentially decreasing road congestion and enhancing safety outcomes.⁴ However, the success of platooning hinges on secure, real-time communication and collaboration among vehicles, which are vulnerable to various security threats.⁵ The reputation model of the platooning system faces the challenges of Sybil attack, collusion attack, whitewashing attack and false data injection which will disrupt the performance of platooning system resulting in life-loss incidents, data damage, and money loss. Furthermore, the data of conventional database-based reputation models is not protected because any external or internal party can exploit the system. The data of the reputation management system is insecure due to lack of security, transparency, and immutability. Refs.^{6, 7} advocates that blockchain should be used for transactions in real time systems to prevent it from external attacks. It has been argued in ref.⁸ that a distributed system is unprotected to a large number of potential threats that increase significantly when distributed systems are open and competitive. So trust and reputation systems of agent's activities and distributed systems can benefit from cryptographic techniques for enhanced security.

In vehicle platooning, maintaining a cohesive and efficient formation is crucial for enhancing traffic flow, reducing fuel consumption, and improving safety. However, the effectiveness of platooning relies heavily on the reliability and cooperation of individual vehicles. Current reputation models for vehicle platooning are inadequate in ensuring consistent and trustworthy behavior among participants, often lacking transparency and resistance to manipulation. Moreover, there is no security mechanism in traditional database based reputation models as any internal or external entity can manipulate such a system.

In order to solve the problem of choosing a reliable agent-vehicle in such a mobile, distributed context and, at the same time, avoiding the adoption of a central repository, we introduce the following ideas.

Modeling how reputation scores of individual vehicles based on their working can be used to ensure that only reliable and trustworthy vehicles are allowed to join a platoon.

Adopting a solution for a distributed reputation management using blockchain technology for transparency, security, and immutability of reputation data for vehicles participating in platooning.

Devising methods for calculation of reputation scores that will be saved in blockchain and used for achieving common global goal.

The main focus of this paper is how to leverage blockchain technology to ensure decentralized and immutable reputation management, aligning individual vehicle actions with the common global goal of optimal platooning performance.

2 Literature review

Ref.⁹ has proposed a reputation based truth discovery model to acquire the truth from conflicting information to quantify long-term quality of sources, aiming to improve the accuracy and reliability. Their work is focused on IoT devices. In ref.¹⁰ a light weight confidentiality preservation scheme with efficient reputation management is proposed to analyze the critical issues of sensing vehicle reliability and achieve light weight confidentiality preservation to select the trustworthy sensing vehicles. Both of these works focused on increasing the efficiency of the platoon without any prevention from attacks.

In ref.,¹¹ a model for estimating reputation and trust retrieving values from blockchain ledger is proposed for e-payment systems, which makes it immutable to data tampering. However, their work does not utilize the multi-agents for decision making.¹² introduced Leadership Incentives for Platoon (LIPs) protocol by using blockchain technology. Their main purpose was to design a payment system that rewards individuals for taking the lead in dynamic platooning with heterogeneous vehicles. Although they used blockchain to prevent the platoon data from tampering, they have no mechanism of increasing the efficiency of a platoon using the reputation scores. In ref.¹³ vehicles use the Bayesian Inference Model to verify the messages they receive from nearby vehicles in a decentralized trust management system for vehicular networks based on blockchain technology. Then joint Proof of Work and Proof of Stake consensus mechanism on Roadside Units (RSUs) add block after rating generated by vehicle for each information vehicle source. Their work is not applicable to the vehicle platooning paradigm as it uses blockchain message storage in a generic vehicle to vehicle communication network.

In ref.¹⁴ a reputation system-based lightweight message protocol and architecture for assessing the credibility of messages in 5G enabled vehicle networks is proposed. Only vehicles with a certain threshold of trust will be able to send messages. Their work did not use the decentralized blockchain storage of messages and the messages are vulnerable to traditional attacks. Ref.¹⁵ has proposed an election scheme and incentive mechanism for the selection of leader in opportunistic autonomous vehicle platoons based on reputation. In this scheme a reputation based election scheme elects a leader who is comparatively trustworthy, in consonance with repute score safely stored in blockchain. Similarly,¹⁶ has discussed a blockchain based approach to enhance trust, security and reliability within Cooperative Adaptive Cruise Control (CACC) platoons. Trusted leader is selected in CACC platoons to improve safety and efficiency and integrated blockchain technology to provide a comprehensive solution for trust and platoon management in CACC-based connected vehicle systems. However, in both of these works, the system does not have agents for autonomous decision making and requires human intervention at various phases.

Refs.^17–21 proposed methods for improving the efficiency of platoons by improving driving decisions with the help of incorporating social intelligence, formally verifying the regulations for the platoons, utilizing reputation scores to assess the reliability of the received data and developing a recommendation system to detect malicious vehicles. But none of these techniques has used blockchain for saving the reputation scores that will prevent the platoons from external attacks and make the reputation scores immutable.

Ref.²² proposed a scheme for evaluating a vehicle's reputation using both historical interactions and recommended opinions from other vehicles. The candidates with high reputation are selected to be active miners and standby miners. A newly generated block of blockchain is verified and audited by standby miners. Ref.²³ presented a reputation management system, capable of identifying malicious actors, to mitigate their effects on the vehicle network. They used a blockchain-based backend for the reputation management system. Ref.²⁴ propose a method that exploits consortium blockchain to maintain transparency and trust in trading activities in the trustless environment of Internet of Electric Vehicles to resolve the trading disputes. Ref.²⁵ has proposed a new blockchain based framework for securing smart vehicles to tailor information access to restricted entities in the connected vehicle ecosystem. They used a challenge–response data exchange between the vehicles to monitor the internal state of the vehicle to identify cases of in-vehicle network compromise. Ref.²⁶ designed a consortium blockchain-based resource sharing paradigm in Internet of Vehicles, in which the resource sharing interactions are encapsulated as transactions and recorded by Road Side Units to cope with the challenge of establishing trust and preserving privacy during the resource sharing process. All of these works are not directly applicable to vehicle platooning where the leaders devise the policies for the platoon.

Ref.²⁴ has used blockchain to enable the payment based on blockchain between the platoon leader and platoon members to avoid the malicious and false payments. Their work does not focus on improving the efficiency of the platoon using the reputations scores. Ref.²⁷ used a reputation-based crowdsourcing framework built on a blockchain platform to support the management of crowdsourcing trading and user-reputation evaluating activities. The reputation values of participants and reveals any malicious behavior accordingly. Their reputation scores are used to provide real-time transportation information so the work is not applicable to vehicle platooning where the focus is to make the platooning data secure and increase the efficiency of the platoon. Ref.²⁸ proposed a blockchain-based trust management model for VANETs to allow vehicles to send messages anonymously in the non-fully trusted environment to guarantee the privacy of the vehicle. The work cannot be extended to vehicle platooning as we do not allow anonymous vehicles to be part of the platoon.

In ref.²⁹ a decentralized trust management system based on blockchain that uses smart contracts to update and maintain trustworthy values throughout the network of connected vehicles is presented but they neither used multi-agents for decentralized decision making nor applicable to vehicle platooning.

None of the related work focused on devising parameters specifically for vehicle platooning to increase the efficiency of the platoon and neither have they utilized the blockchain to store the reputation scores of the vehicles to make the reputation scores immutable from external attacks. In Table 1, we present summarized data about the limitations of existing work.

Table 1.
Summary of the limitations of existing work.

Method/Model Description Limitations

Reputation-Based Truth Discovery framework. Yan et al. (2021)⁹.

PPRM model. Cheng et al (2022).¹⁰
Focuses on efficiency but does not prevent attacks on vehicle platooning. Lack of attack prevention mechanisms.

Deriving Trust and Reputation model. Ahn et el (2019).¹¹

RLE model. Ying et al (2021).¹⁵

Blockchain-integrated Cooperative Adaptive Cruise Control framework. Singh et al (2020).¹⁶

Blockchain-based Decentralized Trust Management framework. Singh et al (2020).²⁹
Requires human intervention at various phases instead of using autonomous agents. Does not use multi-agent systems for autonomous decision-making.

LIPs protocol. Ledbetter et al (2019).¹²

Smart-Contract-Based Economical Platooning model. Chen et al (2019).²⁴
Uses blockchain for security but does not enhance platoon efficiency via reputation scores. No efficiency improvement using reputation scores.

Blockchain-Based Decentralized Trust Management Framework. Yang et al (2018).¹³

Reputation Management Technique for Autonomous Vehicles. Kianersi et al (2022).²³

B-FERL model. Oham et al (2021).²⁵

PoR consensus mechanism. Chai et al (2019).²⁶

RC-Chain model. Sun et al (2021).²⁷

Blockchain-Based Conditional Privacy-Preserving Trust Management Scheme. Liu et al (2019).²⁸
Designed for general vehicular networks or transportation systems but not for platooning. Not applicable to vehicle platooning.

RSMA model . Cui et al (2019).¹⁴
Reputation-based message filtering but lacks blockchain storage, leaving messages open to attacks. No decentralized blockchain storage, making messages vulnerable to attacks.

CPSI model . Zhiwei et al (2022).¹⁷

Double-Level Model Checking Approach. Alves et al (2021).¹⁸

V2X Reputation System with Privacy Preservation. Michalopoulos et al (2019).¹⁹

TPPR method. Zhang et al (2019).²⁰

EMPC method. Hu et al (2022).²¹
Uses different methods to improve platooning but does not leverage blockchain for reputation score immutability. Does not use blockchain for reputation score storage.

Reputation and Contract-Theory-Based Consensus Management Scheme. Kang et al (2019).²²

Smart-Contract-Based Economical Platooning Model. Chen et al (2019).²⁴
Addresses aspects like blockchain miner selection and payments but does not tackle platooning efficiency. Focuses on specific blockchain applications not directly related to platooning.

Method/Model	Description	Limitations
Reputation-Based Truth Discovery framework. Yan et al. (2021)⁹. PPRM model. Cheng et al (2022).¹⁰	Focuses on efficiency but does not prevent attacks on vehicle platooning.	Lack of attack prevention mechanisms.
Deriving Trust and Reputation model. Ahn et el (2019).¹¹ RLE model. Ying et al (2021).¹⁵ Blockchain-integrated Cooperative Adaptive Cruise Control framework. Singh et al (2020).¹⁶ Blockchain-based Decentralized Trust Management framework. Singh et al (2020).²⁹	Requires human intervention at various phases instead of using autonomous agents.	Does not use multi-agent systems for autonomous decision-making.
LIPs protocol. Ledbetter et al (2019).¹² Smart-Contract-Based Economical Platooning model. Chen et al (2019).²⁴	Uses blockchain for security but does not enhance platoon efficiency via reputation scores.	No efficiency improvement using reputation scores.
Blockchain-Based Decentralized Trust Management Framework. Yang et al (2018).¹³ Reputation Management Technique for Autonomous Vehicles. Kianersi et al (2022).²³ B-FERL model. Oham et al (2021).²⁵ PoR consensus mechanism. Chai et al (2019).²⁶ RC-Chain model. Sun et al (2021).²⁷ Blockchain-Based Conditional Privacy-Preserving Trust Management Scheme. Liu et al (2019).²⁸	Designed for general vehicular networks or transportation systems but not for platooning.	Not applicable to vehicle platooning.
RSMA model . Cui et al (2019).¹⁴	Reputation-based message filtering but lacks blockchain storage, leaving messages open to attacks.	No decentralized blockchain storage, making messages vulnerable to attacks.
CPSI model . Zhiwei et al (2022).¹⁷ Double-Level Model Checking Approach. Alves et al (2021).¹⁸ V2X Reputation System with Privacy Preservation. Michalopoulos et al (2019).¹⁹ TPPR method. Zhang et al (2019).²⁰ EMPC method. Hu et al (2022).²¹	Uses different methods to improve platooning but does not leverage blockchain for reputation score immutability.	Does not use blockchain for reputation score storage.
Reputation and Contract-Theory-Based Consensus Management Scheme. Kang et al (2019).²² Smart-Contract-Based Economical Platooning Model. Chen et al (2019).²⁴	Addresses aspects like blockchain miner selection and payments but does not tackle platooning efficiency.	Focuses on specific blockchain applications not directly related to platooning.

3 Proposed blockchain-based reputation model for vehicle platooning framework

Our proposed framework, Blockchain-based Reputation Model for Vehicle Platooning (BRMVP) is shown in Figure 1. The framework makes use of smart contract and real-time calculation of reputation scores to achieve a common global goal for vehicle platoons.

Figure 1.

Proposed Blockchain-based Reputation Model for Vehicle Platooning Framework.

3.1 Platoon agent

On every vehicle, there will be an agent deployed which is named as Platoon agent. We have divided its working into two components. One will be a smart contract that will handle all the data related to reputations in the form of a smart contract. The second one is the Managed Component which will receive the data required for the calculation of reputations. It is to be noted that the communication of a platoon agent with other agents will be using the standard FIPA performatives. For every platoon member, there will be an agent deployed, which will be responsible for storing the data related to the reputation of all the agents it has interacted with. Additionally, it will maintain the data of computed reputations in a smart contract.

The data has been named as Common Global Goal (CGG) in our system, self-executing contracts known as smart contracts have the conditions of the agreement between all the agents. This means that when an agent has computed its reputation for an agent the entry will be stored in all the agents that are currently part of the system. This will ensure that no agent or an attacker is able to compromise the reputation scores of an agent. The smart contracts will be executed and enforced by a blockchain network. These contracts will be triggered whenever a vehicle agent is about to make a decision for allowing a vehicle to be part of the platoon or a vehicle wants to become a part of the platoon. Similarly when a vehicle wants to adjust its position in a platoon as a result of messages sent by the leader.

This component is responsible for managing the data required for the computation of reputations. In our system we have devised the parameters of Velocity, Distance, Ping signal, and Reliability. These values will be continuously retrieved by an agent so that when a decision needs to be made, the average of these values can be used. In our system whenever there will be a change required in the dynamic of the system it will be considered as an event. These corresponds to

Event 1: Allowing a vehicle to join the platoon

Event 2: A vehicle is leaving the platoon

Event 3: Forcefully removing a vehicle from the platoon due to not following the instructions

Event 4: Rearrangement of a platoon for efficiency

3.2 Leader agent

In every platoon, there will be an agent deployed on the head vehicle known as Leader agent. It will devise all the rules related to the calculation of reputations. Any vehicle who wants to join the platoon will have to ask permission from the leader agent. Similarly, any vehicle who wants to leave the platoon will also have to ask the leader agent. It will also be responsible for starting and stopping the platoon. In addition, this agent will be managing the communication with all the other vehicles.

4 Reputation model for vehicle platooning

Now we define the reputation model for an agent-based vehicle to improve the effectiveness of platooning by only allowing those agents to be part of the platoon which meets certain criteria. With this model, vehicles will be able to determine the trustworthiness of their cooperators and eliminate the unknown and possibly malicious partners which are creating safety hazards. The parameters of our proposed reputation model include:

4.1 Velocity

The velocity of the followers cannot exceed the leader's velocity in the platoon.

\begin{aligned} V_{i} = {1 - \frac{v_{l} - v_{f}}{v_{l}}} \end{aligned}

(1)

where

v_{l}

denotes the velocity of the leader and

v_{i}

is the velocity of the i^th vehicle. The range of formula is [0,1]. Where higher value (close to 1) indicates more accurate and well-matched data. While lower value (close to 0) indicates discrepancies or error.

4.2 Distance

The intra-vehicular distance must be fixed in the platoon. Each vehicle must maintain the predefined distance fixed by the leader in the platoon in order to maintain the safety. The distance can be calculated using the gps locations of the vehicles.

\begin{aligned} D_{i} = {\begin{matrix} 1 i f l o c_{i - 1} - l o c_{i} = 5 \\ 0 i f l o c_{i - 1} - l o c_{i} > 5 \end{matrix} \end{aligned}

(2)

Where $l o c_{i - 1}$ represents the gps location of the predecessor vehicle of i^th vehicle and $l o c_{i}$ is the gps location of the i^th vehicle. The intra-vehicular distance should be equal to the safe distance threshold which is 5 m. Each vehicle will be responsible for maintaining the distance from its front vehicle except for the leader.

4.3 Ping signal

In platooning, ping signals play a crucial role in maintaining communication, as communication is the big critical segment of platooning. It indicates the responsiveness of a vehicle in a platoon. Every vehicle periodically sends signals to the leader to maintain the state of their connections. After every 30 s, each vehicle is supposed to send a signal to the leader as its heartbeat.

\begin{aligned} P_{i} = {\begin{matrix} 1 i f P_{l} = P_{f} \\ \frac{P_{l}}{P_{f}} i f P_{l} < P_{f} \end{matrix} \end{aligned}

(3)

Where $P_{i}$ denotes the ping signal value for the i^th vehicle. $P_{l}$ are the ping signals received by the leader and $P_{f}$ are the ping signals sent by the follower vehicle.

4.4 Reliability

The historical behavior of a vehicle, including its past performance in the platooning scenarios, contributes to its reliability. Vehicles with a positive track record will be considered more reliable in a platoon. The reliability of a vehicle refers to its dependability, trustworthiness, and the likelihood that it will perform as expected within the platoon. The reliability of a vehicle in the platooning system is crucial for the safety and efficiency of the overall system's operation.

\begin{aligned} R_{i} = {\begin{cases} 1, & i f \frac{N E}{T} = 0 \\ 1 - \frac{N E}{T}, & i f \frac{N E}{T} < 0.5 \\ 0, & i f \frac{N E}{T} \geq 0.5 \end{cases} \end{aligned}

(4)

Where NE represents the total number of malfeasance interactions of an agent-based vehicle and T represents the total number of interactions of that vehicle within the platoon. The range of this function is [0,1], zero will be assigned to the vehicle exhibiting malicious activity in the platoon. This attribute can only be computed for a vehicle that has joined the platoon in the past.

Reputation in terms of Utility Function can be defined as

\begin{aligned} (U_{i}) = {V_{i} + D_{i} + P_{i} + R_{i}} \end{aligned}

(5)

The range of this function for a single agent-based vehicle is [0,4]. After computation it will be decided by the leader that which vehicle will be given priority to become part of the platoon.

4.5 Objectives

The system designer can alter these objectives as per the need.

(a)
Maximize Communication Efficiency: Ensure that agents can exchange information quickly and reliably. We can define this objective in terms of the weighted utility function. For effective communication, ping signal has the most value so we will assigns 50% weight to the ping signal. Velocity and distance are given 15% each, contributing equally. Reliability gets 20% weight, which is also significant.
$Objective = max ((0.50 \times P) + (0.15 \times V) + (0.15 \times D) + (0.20 \times R))$

(b)
Maximize Platoon Speed: Platoon should travel at the best possible speed. In this case we will define the objective as max optimization problem by giving more weights to velocity and distance.
$Objective = max ((0.20 \times P) + (0.30 \times V) + (0.30 \times D) + (0.20 \times R))$

(c)
Ensure Robustness: Improve the system's ability to handle disruptions or failures. In this case a vehicles historical reliability means more to us than its current behavior so it will be given more weight.
$Objective = max ((0.20 \times P) + (0.15 \times V) + (0.15 \times D) + (0.50 \times R))$

5 Smart contract for blockchain-based reputation model for vehicle platooning framework

We have created the following smart contract for saving the values of Velocity, Distance, Ping signal and Reputation in blockchain. It is designed in such a way that there will be one to many entries of the computed values. This means that an agent will compute and save the values of all of its neighbours. These values will be used for runtime computation of reputation score whenever there will be such an event. This contract allows for the storage and retrieval of crucial data points relevant to vehicle platooning, facilitating efficient communication and data management within the system. The contract is written in Solidity Language and will be executed on the Ethereum blockchain. The complete code is presented in Algorithm 1. We used Remix IDE for the compilation of our smart contract.

Algorithm 1.
Solidity code for smart contract execution

1. contract VehiclePlatooning {

2. // Struct to store vehicle data

3. struct VehicleData {

4. string vehicleName;

5. int256 distance;

6. int256 velocity;

7. int256 pingSignal;

8. int256 reliability;

9. }

10. // Mapping to store data for each neighboring vehicle by the owner's address

11. mapping(address => mapping(address => VehicleData)) public vehicleData;

12. // Events

13. event DataUpdated(address indexed owner, address indexed neighbor, string vehicleName, int256 distance, int256 velocity, int256 pingSignal, int256 reliability);

14. // Function to set data for a neighboring vehicle

15. function setDataForNeighbor(address _neighbor, string memory _vehicleName, int256 _distance, int256 _velocity, int256 _pingSignal, int256 _reliability) public {

16. VehicleData storage data = vehicleData[msg.sender][_neighbor];

17. data.vehicleName = _vehicleName;

18. data.distance = _distance;

19. data.velocity = _velocity;

20. data.pingSignal = _pingSignal;

21. data.reliability = _reliability;

22. emit DataUpdated(msg.sender, _neighbor, _vehicleName, _distance, _velocity, _pingSignal, _reliability);

23. }

24. // Function to get data for a neighboring vehicle

25. function getDataForNeighbor(address _owner, address _neighbor) public view returns (string memory vehicleName, int256 distance, int256 velocity, int256 pingSignal, int256 reliability) {

26. VehicleData storage data = vehicleData[_owner][_neighbor];

27. return (data.vehicleName, data.distance, data.velocity, data.pingSignal, data.reliability);

28. }

29. }

Figure 2.

A sample platoon with optimized working based on reputations.

6 Application of the proposed blockchain-based reputation model for vehicle platooning framework

In this section we have shown the working of the proposed BRMVP framework with the help of a case study. Figure 2 shows a platoon in which there are four vehicles named V1, V2, V3, V4 that are part of a platoon with a leader. We assume that the capacity of the vehicle is at most fives to maximize the efficiency of the platoon as less communication overhead will occur. Three more vehicles want to join the platoon but there is only a capacity for one. The goal is how the reputation model can be utilized so that we allow the vehicle that will increase the performance of the overall platoon. This will be in contrast to allowing any random vehicle to be part of the platoon or allowing them on a first come first serve basis.

Let's summarize the data of the above platoon in Table 2 so that we can apply the reputation model.

Table 2.
Summary of attributes of all the vehicles.

Vehicle Velocity Distance PingSignal Reliability

Leader 60 Kmh NA NA NA

V1 58 Kmh 2 m 1 X

V2 58 Kmh 3 m 0.6 1

V3 57 Kmh 4 m 0.5 X

V4 59 Kmh 5.5 m 0.8 0.7

V5 X X X 0.8

V6 X X X X

V7 X X X 0.9

Vehicle	Velocity	Distance	PingSignal	Reliability
Leader	60 Kmh	NA	NA	NA
V1	58 Kmh	2 m	1	X
V2	58 Kmh	3 m	0.6	1
V3	57 Kmh	4 m	0.5	X
V4	59 Kmh	5.5 m	0.8	0.7
V5	X	X	X	0.8
V6	X	X	X	X
V7	X	X	X	0.9

Where X corresponds to data that is required but currently is not available at the moment. We assume that we have reputation scores for vehicle V5 and V7 only since they joined the platoon previously. Our platoon has the policy that if there are limited slots in the platoon and more vehicles want to join the platoon then the decision will favour those vehicles that joined the platoon in the past and has the highest reputation score. Now that we have the data required for calculating reputations we can compute them as shown below.

6.1 Calculation of reputation scores

\begin{matrix} Re p_{V 1} = {V_{1} + D_{1} + P_{1} + R_{1}} = 1 - (\frac{(60 - 58)}{60}) + 1 + 1 + 0 = 0.966 + 1 + 1 = 2.966 \end{matrix}

(6)

\begin{matrix} Re p_{V 2} = {V_{2} + D_{2} + P_{2} + R_{2}} = 1 - (\frac{(60 - 58)}{60}) + 1 + 0.6 + 1 = 0.966 + 1 + 0.6 + 1 = 3.566 \end{matrix}

(7)

\begin{matrix} Re p_{V 3} = {V_{3} + D_{3} + P_{3} + R_{3}} = 1 - (\frac{(60 - 57)}{60}) + 1 + 0.5 + 0 = 0.95 + 1 + 0.5 + 0 = 2.45 \end{matrix}

(8)

\begin{matrix} Re p_{V 4} = {V_{4} + D_{4} + P_{4} + R_{4}} = 1 - (\frac{(60 - 59)}{60}) + 0 + 0.8 + 0.7 = 0.98 + 0 + 0.8 + 0.7 = 2.48 \end{matrix}

(9)

\begin{matrix} Re p_{V 5} = {V_{5} + D_{5} + P_{5} + R_{5}} = 0.8 \end{matrix}

(10)

\begin{matrix} Re p_{V 7} = {V_{7} + D_{7} + P_{7} + R_{7}} = 0.9 \end{matrix}

(11)

Equations 6–11 show the computed scores for the reputations. Using the smart contract we stored these reputation scores in a private blockchain.

Figure 3.

Leader block saved in blockchain.

6.2 Using blockchain to save reputation scores

Since leader does not follow any vehicle, we have assigned it default values of 1 for all the attributes required for calculating reputations. The Figure 3 shows how the leader block is saved in the blockchain.

Mining of V1 block is shown in the Figure 4.

Mining of V2 block is shown in Figure 5.

Mining of V3 block is shown in Figure 6.

Mining of V4 block is shown in Figure 7.

Figure 4.

Vehicle 1 block saved in blockchain.

Figure 5.

Vehicle 2 block saved in blockchain.

Figure 6.

Vehicle 3 block saved in blockchain.

Figure 7.

Vehicle 4 block saved in blockchain.

6.3 Defining common global goal

We discussed the different strategies for achieving common global goals. For our case study we chose the common global goal of Adaptive Velocity Control. This common global goal requires the platoon to implement the features of spacing adjustment, collision avoidance, and traffic flow optimization. Our goal is defined in Eq 12.

\begin{aligned} max {Average Fuel Comsumption of Platoon} \end{aligned}

(12)

6.4 Simulation of platoon using anylogic

We have created the platoon as described in our case study to show how to achieve the common global goal and the effect of using reputation based platoon on the platoon performance. The environment is created in Anylogic using the agent based discrete event simulation. The speed of vehicles is set at 50 kms per second. The model creates and agent instance randomly that can be access in the code using its identifier. We set the execution mode as real time with scale of 1. Figure 8 shows the platoon in which there is one leader agent and four vehicles named as V1, V2, V3 and V4. There are three type of agents in our model named vehicle agent, leader agent and a main agent. The vehicle agent will represent each vehicle on the road. These vehicles can be part of the platoon or not.

Figure 8.

Platoon created in Anylogic.

The Figure 9 below shows the working of a vehicle agent. A vehicle agent will be in normal state until it will receive a message to become part of the platoon. The initial speed of a vehicle has been set at 10 meters per second.

Figure 9.

Statechart of a vehicle agent.

The main agent will contain two types of agents namely vehicle agent and leader agent. We will maintain an array list of vehicle agents as there can be many vehicles in a platoon. The complete background of the simulation is designed in the presentation of the main agent. The process of vehicle creation is controlled by an event named createVehicle with trigger type of timeout and mode of cyclic. Its recurrence time is set from 0.5 to 0.7 s. We have also defined a function platooningFunction that will display the vehicle labels for vehicles that are part of the platoon.

The Figure 10 shows the scenario in which 3 more vehicles want to join the platoon and V6 does not have interaction with the platoon previously. As there is only one more space left in the platoon for any new vehicle, the leader has to decide which vehicle should be allowed.

Figure 10.

New vehicles want to join the platoon.

We computed the reputation scores for all the vehicles. This also included the three new vehicles that wanted to be part of the platoon. Based on that score the leader will allow V7 to be part of the platoon as it has high reliability value as compared to V5. The Figure 11 demonstrates this use case.

Figure 11.

Vehicle V7 has joined the platoon.

Now we need to decide how to calculate the efficiency of the platoon before the vehicle V7 was allowed to be part of the platoon and after it became its part. We shall compute it by calculating the average fuel consumption of the whole platoon. This was set as a common global goal of the platoon in the previous section. The Figure 12 below shows the average fuel consumption before the vehicle V7 was part of the platoon. The average fuel consumption is 12/km/liter. The word FC stands for fuel consumption.

Figure 12.

Average fuel consumption of the platoon.

The graph integrated in Figure 12 shows the average fuel consumption of vehicles.

We will now see the effect of allowing V7 to be part of the platoon on average fuel consumption.

After the vehicle V7 joined the platoon, the average fuel consumption is 12.33 km/liter as shown in Figure 13.

Figure 13.

Average fuel consumption of the platoon after V7 joined it.

7 Discussion

The Table 3 provides a comparative analysis highlighting key benefits of using the blockchain-based reputation model for vehicle platooning with a common global goal.

Table 3.
Comparative analysis of the proposed approach.

Traditional Approaches BRMPV Approach

Trust and Transparency 1.
Often rely on centralized authorities or trusted third parties to manage and verify reputations.
2.
Prone to single points of failure and can be susceptible to tampering and corruption.
3.
Lack transparency, as the decision-making processes and reputation calculations are not always open to public scrutiny.
1.
Utilizes a decentralized ledger that eliminates the need for a central authority, thereby reducing the risk of single points of failure.
2.
Ensures transparency and immutability, as all transactions and reputations are recorded on a publicly verifiable blockchain.
3.
Enhances trust among participants by providing a tamper-proof and transparent record of vehicular interactions.

Security and Privacy 1.
Vulnerable to various security threats, including data breaches, spoofing, and Sybil attacks.
2.
May not adequately protect the privacy of vehicles and drivers, as centralized systems can become targets for data harvesting.
1.
Employs cryptographic techniques to secure data, ensuring that only authorized parties can access sensitive information.
2.
Protects against common security threats by using distributed consensus mechanisms and cryptographic hashing.
3.
Maintains the privacy of vehicle identities while allowing for the verification of reputations, thus providing a balance between transparency and privacy.

Reliability and Robustness 1.
Centralized systems are more susceptible to disruptions and failures, which can compromise the reliability of the entire platooning network.
2.
Trust and reputation scores may be manipulated, reducing the overall robustness of the system.
1.
Distributed nature ensures high availability and fault tolerance, as the network can continue to operate even if some nodes fail.
2.
Immutable and verifiable records prevent the manipulation of reputation scores, thus enhancing the robustness of the system.
3.
The consensus mechanism ensures that all participants agree on the state of the network, further improving reliability.

Scalability and Performance 1.
May face scalability issues as the number of participating vehicles increases, leading to performance bottlenecks.
2.
Centralized systems often struggle to handle a large volume of transactions efficiently.
1.
Designed to scale effectively with the number of participating vehicles, as the workload is distributed across the network.
2.
Can handle a high volume of transactions with minimal latency, thanks to efficient consensus algorithms and optimized data structures.
3.
The use of smart contracts can automate and streamline various processes, further enhancing performance.

	Traditional Approaches	BRMPV Approach
Trust and Transparency	1. Often rely on centralized authorities or trusted third parties to manage and verify reputations. 2. Prone to single points of failure and can be susceptible to tampering and corruption. 3. Lack transparency, as the decision-making processes and reputation calculations are not always open to public scrutiny.	1. Utilizes a decentralized ledger that eliminates the need for a central authority, thereby reducing the risk of single points of failure. 2. Ensures transparency and immutability, as all transactions and reputations are recorded on a publicly verifiable blockchain. 3. Enhances trust among participants by providing a tamper-proof and transparent record of vehicular interactions.
Security and Privacy	1. Vulnerable to various security threats, including data breaches, spoofing, and Sybil attacks. 2. May not adequately protect the privacy of vehicles and drivers, as centralized systems can become targets for data harvesting.	1. Employs cryptographic techniques to secure data, ensuring that only authorized parties can access sensitive information. 2. Protects against common security threats by using distributed consensus mechanisms and cryptographic hashing. 3. Maintains the privacy of vehicle identities while allowing for the verification of reputations, thus providing a balance between transparency and privacy.
Reliability and Robustness	1. Centralized systems are more susceptible to disruptions and failures, which can compromise the reliability of the entire platooning network. 2. Trust and reputation scores may be manipulated, reducing the overall robustness of the system.	1. Distributed nature ensures high availability and fault tolerance, as the network can continue to operate even if some nodes fail. 2. Immutable and verifiable records prevent the manipulation of reputation scores, thus enhancing the robustness of the system. 3. The consensus mechanism ensures that all participants agree on the state of the network, further improving reliability.
Scalability and Performance	1. May face scalability issues as the number of participating vehicles increases, leading to performance bottlenecks. 2. Centralized systems often struggle to handle a large volume of transactions efficiently.	1. Designed to scale effectively with the number of participating vehicles, as the workload is distributed across the network. 2. Can handle a high volume of transactions with minimal latency, thanks to efficient consensus algorithms and optimized data structures. 3. The use of smart contracts can automate and streamline various processes, further enhancing performance.

From the results of the simulations and the table above we can see that using the dynamic attributes of the platoon members i.e., velocity, distance, ping signal and reliability will lead to overall increased performance of the system. The usage of blockchain additionally prevents the system from centralized failure and attacks. Not to mention that the proposed work can effectively be used in holonic multi agent systems, where multiple distributed platoons can communicate with each other to achieve the global goal without facing any scalability issues.

8 Conclusion

In this research we proposed a framework for the integration of blockchain technology into the development of a reputation model for vehicle platooning systems with a common global goal. The primary aim was to enhance the reliability, security, and efficiency of vehicle platooning by leveraging the decentralized and immutable nature of blockchain. We designed a blockchain-based reputation model with four attributes that ensures transparent and tamper-proof recording of vehicular interactions. All the agents store their reputation scores using smart contracts in a blockchain. This ensures that no agent can compromise any other agent's score as all share a common ledger. Establishing trust like this among agents ensures that the platoon performance is optimized. We demonstrated the application of our proposed framework using a case study using the maximization of average fuel consumption as a common global goal. The results show that using the four attributes for making decisions regarding which vehicles should be part of the platoon will help to achieve the common global goal.

Currently, only the agent's data (reputation scores) are stored in a blockchain using smart contracts. If we want to establish complete trust among agents of a multi-agent system then the complete messages exchanged between agents should also be saved in a blockchain. For this we need to ensure the messages are communicated in a standard format.

Footnotes

ORCID iDs

Hanaa Nafea

Awais Qasim

Attiya Hussain

Ilyas Fakhir

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Author Biographies

Hanaa Nafea received the master's degree in computer science from Nottingham Trent University (NTU), Nottingham, U.K., in 2012. She received the Ph.D. degree with Liverpool John Moores University (LJMU), U.K. in 2020. She is also assistant professor at Taibah University, Al-Madinah, Al-Munawwarah, Saudi Arabia. Her current research interests include network security, the security of complex systems, intrusion detection, secure service composition, privacy-preserving data aggregation, cryptography computer science, and cloud security.

Awais Qasim received the M.S. degree in Computer Science from Lahore University of Management Sciences (LUMS), Lahore, Pakistan, in 2011. After that, he worked as a Software Engineer in Industry and developed a number of iPhone and Android applications. After that he completed his Ph.D in Computer Science in 2017. Then we completed Post-doc from University of Salford in 2020. Currently, he is working as Assistant Professor in the Computer Science Department, Government College University. His research interests include model checking, multi-agent systems, real-time systems, self-adaptive systems, and robotics for blockchain, and disaster resilience. He is the author of 36 research articles.

Attiya Hussain is currently serving as a Lecturer in the Department of Computer Science at Green International University (GIU). She holds a Master of Science degree in Computer Science from Government College University (GCU), Lahore. Prior to joining GIU, she worked as a Computer Science faculty member at USA Cantt Campus. Her research interests lie in blockchain technology, intelligent transportation systems, and trust-based reputation models in vehicular networks. Attiya is committed to academic excellence, student mentorship, and innovative research in emerging technological domains.

Ilyas Fakhir received PhD degree in Computer Science from GC University, Lahore in 2019. He did the M.Sc degree in Computer Science from the Iqra University, Karachi, Pakistan, in 2003, and the MS in Computer Science from the University of Central Punjab, Lahore, in 2012. He joined the Department Computer Science, Government College University, in 2008 as a Lecturer and presently working as an Assistant Professor. He has authored 20 research papers in peer-reviewed ISI indexed journals. His research areas include Petri Nets, multi-agent systems, Concurrent systems, and self-adaptive systems.

References

Maldonado

Cruz

Torres

, et al. “Multi-agent systems: a survey about its components, framework and workflow. IEEE Access 2024; 12: 2169–3536.

Razzaghpour

Soorchaei

Valiente

, et al. Mass platooning: information networking structures for long platoons of connected vehicles. IEEE Open J Intell Transp Syst 2024; 5: 2687–7813.

Deng

Liu

, et al. “Towards safe and robust autonomous vehicle platooning: A self-organizing cooperative control framework. 2408.09468. 2024.

Dokur

Katkoori

. Internet of vehicles-based autonomous vehicle platooning. SN Comput Sci 2023; 1: 2661–8907.

Han

Q-L

, et al. Resilient and safe platooning control of connected automated vehicles against intermittent denial-of-service attacks. IEEE/CAA J Autom Sin 2022; 5: 1234–1251.

Qasim

Bilal

Munawar

, et al. Blockchain based intrusion detection in agent-driven flight operations. Multiagent Grid Syst 2024; 20: 161–183.

Qasim

Ghouri

Munawar

. An effective approach for reducing data redundancy in multi-agent system communication. Multiagent Grid Syst 2024; 20: 69–88.

Kumar

MGV

Ragupathy

. A survey on current key issues and status in cryptography. 2016.

Yan

Yang

. Reputation-based truth discovery with long-term quality of source in internet of things. IEEE Internet Things J 2021; 7: 5410–5421.

10.

Cheng

Liu

, et al. A lightweight privacy preservation scheme with efficient reputation management for mobile crowdsensing in vehicular networks. IEEE Trans Dependable Secure Comput 2022; 3: 1771–1788.

11.

Ahn

Park

Shin

, et al. A model for deriving trust and reputation on blockchain-based e-payment system. Appl Sci 2019; 9: 5362.

12.

Ledbetter

Wehunt

Rahman

, et al. LIPs: a protocol for leadership incentives for heterogeneous and dynamic platoons. 2019 IEEE 43rd Annual Computer Software and Applications Conference (COMPSAC) 2019; 1: 535–544.

13.

Yang

Lei

, et al. Blockchain-based decentralized trust management in vehicular networks. IEEE Int Things J 2018; 6: 1495–1505.

14.

Cui

Zhang

Zhong

, et al. RSMA: reputation system-based lightweight message authentication framework and protocol for 5G-enabled vehicular networks. IEEE Internet Things J 2019; 6: 6417–6428.

15.

Ying

Zhao

, et al. A reputation-based leader election scheme for opportunistic autonomous vehicle platoon. IEEE Trans Veh Technol 2021; 71: 3519–3532.

16.

Singh

Nandi

, et al. Integrating blockchain with cacc for trust and platoon management. Cryptocurrencies Blockchain Technol Appl 2020: 77–97.

17.

Guo

Meng

Chakraborty

, et al. Autonomous behavioral decision for vehicular agents based on cyber-physical social intelligence. IEEE Trans Comput Soc Syst 2022; 10: 2111–2122.

18.

Alves

Dennis

Fisher

. A double-level model checking approach for an agent-based autonomous vehicle and road junction regulations. J Sens Actuator Netw 2021; 10: 41.

19.

Michalopoulos

Meijers

Singh

, et al. “A V2X reputation system with privacy considerations,” 2022 IEEE 13th International Conference on Software Engineering and Service Science (ICSESS), 8–14, 2022.

20.

Zhang

Zhu

, et al. TPPR: a trust-based and privacy-preserving platoon recommendation scheme in VANET. IEEE Trans Serv Comput 2019; 15: 806–818.

21.

Bian

, et al. Fuel economy-oriented vehicle platoon control using economic model predictive control. IEEE Trans Intell Transp Syst 2022; 23: 20836–20849.

22.

Kang

Xiong

Niyato

, et al. Toward secure blockchain-enabled internet of vehicles: optimizing consensus management using reputation and contract theory. IEEE Trans Veh Technol 2019; 68: 2906–2920.

23.

Kianersi

Uppalapati

Bansal

, et al. Evaluation of a reputation management technique for autonomous vehicles. Fut Int 2022; 14: 31.

24.

Chen

Xiao

Qiu

, et al. Smart-contract-based economical platooning in blockchain-enabled urban internet of vehicles. IEEE Trans Ind Inform 2019; 16: 4122–4133.

25.

Oham

Michelin

Jurdak

, et al. B-FERL: blockchain based framework for securing smart vehicles. Inf Process Manag 2021; 58: 102426.

26.

Chai

Leng

Zhang

, et al. Proof-of-reputation based-consortium blockchain for trust resource sharing in internet of vehicles. IEEE Access 2019; 7: 175744–175757.

27.

Sun

Yang

Chen

, et al. RC-chain: reputation-based crowdsourcing blockchain for vehicular networks. J Netw Comput Appl 2021; 176: 102956.

28.

Liu

Huang

Xiao

, et al. A blockchain-based trust management with conditional privacy-preserving announcement scheme for VANETs. IEEE Internet Things J 2019; 7: 4101–4112.

29.

Singh

Nandi

, et al. Blockchain-based adaptive trust management in internet of vehicles using smart contract. IEEE Trans Intell Transp Syst 2020; 22: 3616–3630.

Blockchain-based reputation model for vehicle platooning with common global goal

Abstract

Keywords

1 Introduction

2 Literature review

3.2 Leader agent

4 Reputation model for vehicle platooning

4.1 Velocity

Table 2. Summary of attributes of all the vehicles. Vehicle Velocity Distance PingSignal Reliability Leader 60 Kmh NA NA NA V1 58 Kmh 2 m 1 X V2 58 Kmh 3 m 0.6 1 V3 57 Kmh 4 m 0.5 X V4 59 Kmh 5.5 m 0.8 0.7 V5 X X X 0.8 V6 X X X X V7 X X X 0.9

Footnotes

ORCID iDs

Funding

Declaration of conflicting interests

Author Biographies

References

Table 2.
Summary of attributes of all the vehicles.

Vehicle Velocity Distance PingSignal Reliability

Leader 60 Kmh NA NA NA

V1 58 Kmh 2 m 1 X

V2 58 Kmh 3 m 0.6 1

V3 57 Kmh 4 m 0.5 X

V4 59 Kmh 5.5 m 0.8 0.7

V5 X X X 0.8

V6 X X X X

V7 X X X 0.9