Secure Communication and Dynamic Formation Control of Intelligent Drone Swarms Using Blockchain Technology

Abstract

With the increasing deployment of unmanned aerial vehicle (UAV) swarms in scenarios such as disaster response, environmental monitoring, and military reconnaissance, the need for secure and scalable formation control has become critical. Traditional centralized architectures face challenges such as limited scalability, communication bottlenecks, and single points of failure in large-scale swarm coordination. To address these issues, this paper proposes a blockchain-based decentralized formation control framework that integrates smart contracts to manage UAV registration, identity authentication, formation assignment, and positional coordination. The system follows a leader–follower structure, where the leader broadcasts formation tasks via on-chain events, while followers respond in real-time through event-driven mechanisms. A parameterized control model based on dynamic angle and distance adjustments is employed to support various formations, including V-shape, line, and circular configurations. The transformation from relative to geographic positions is achieved using Haversine and Euclidean methods. Experimental validation in a simulated environment demonstrates that the proposed method achieves lower communication latency and better responsiveness compared to polling-based schemes, while offering enhanced scalability and robustness. This work provides a feasible and secure decentralized control solution for future UAV swarm systems.

1. Introduction

In recent years, the rapid development of drone technology and the continuous advancement of intelligent algorithms have greatly promoted the application of drone swarms in military reconnaissance, environmental monitoring, emergency rescue, logistics and distribution, and other fields. Inspired by the natural swarming behaviors of bird flocks, fish schools, and insects, researchers are working to ensure that drone swarms have the ability to make autonomous decisions and coordinate operations [1]. With the help of advanced sensors, communication technologies, and intelligent algorithms, drone swarms can achieve efficient and autonomous task planning and execution without human intervention, thereby showing a higher efficiency and adaptability in complex and dynamic environments. In practical applications, drone swarms usually adopt a variety of formation modes, such as straight formation, V formation, diamond formation, and grid formation [2]. These formations not only optimize flight trajectories and reduce energy consumption but also improve the accuracy and stability of task completion. The core of drone formation control technology lies in the real-time adjustment and precise coordination of the relative positions between members. Through advanced guidance algorithms, each drone can flexibly respond to external interference and environmental changes while maintaining the predetermined geometric structure.

However, to achieve efficient coordination, these swarms must rely on internal communication for real-time data exchange and decision-making. Factors such as natural obstacles, insufficient network coverage, and cyber attacks can all interfere with communication links, thereby introducing security risks. Furthermore, as the number of drones increases, both the control complexity and communication overhead grow significantly, resulting in poor adaptability to dynamic environments within large-scale drone formations.

Blockchain technology provides innovative solutions to the above problems. Its core advantages are reflected in the following three aspects:

• Distributed Architecture: The decentralized communication framework inherently adapts to large-scale distributed networks, where nodes propagate block updates via P2P broadcasting, eliminating frequent requests to a central controller [3].
• Data Integrity Assurance: When drones perform tasks, they need to collect and transmit a large amount of data. The hash chain storage structure of the blockchain ensures the integrity of the data during transmission and storage, preventing malicious tampering and forgery [4].
• Smart Contract Autonomy: Smart contracts are automated execution protocols on the blockchain that can automatically trigger corresponding operations when preset conditions are met. In drone clusters, the introduction of smart contracts can achieve autonomous collaboration and task allocation, reduce human intervention, and improve task execution efficiency.

The main motivation of this paper is to combine blockchain technology with the dynamic formation control and secure communication of drone swarms. The purpose of this combination is to develop a more secure and smooth information exchange scheme. This scheme aims to protect the drone swarm from common malicious attacks such as node tampering and deception when transmitting data, ensuring the stability and robustness of drones when they are in dynamic formation. Our contributions are summarized as follows.

(1)

We proposed a blockchain event-driven distributed UAV formation control architecture, which utilizes smart contracts to achieve autonomous UAV registration, task distribution, and formation management. By leveraging the event broadcasting mechanisms, follower UAVs can monitor and respond to formation commands in real-time, effectively addressing the large-scale communication bottleneck issues inherent in traditional control architectures.

(2)

We designed a parametric dynamic formation control algorithm with a conflict resolution mechanism, introducing a mathematical model based on angle–distance dual parameter sets that supports adaptive switching among different formations. The algorithm incorporates Haversine spherical coordinate transformation for large-scale precise positioning and implements a blockchain-based position conflict resolution mechanism, ensuring collision-free formation reconfiguration through Euclidean distance calculation and on-chain position locking.

(3)

We established the first integrated blockchain–flight control simulation verification platform and completed scalability validation. The platform combines Ganache blockchain, PX4 flight control, and Gazebo physics engine to achieve full-process simulation from command issuance to physical execution.

The rest of this paper is organized as follows. Section 2 discusses related work and Section 3 introduces our proposed framework. Section 4 is dedicated to our implementation. In Section 5, we present simulation results for drone swarm formations and performance analysis. Section 6 summarizes our findings and provides suggestions for future research.

2. Related Work

2.1. Formation Control

The current research on drone formation control mainly focuses on coordination, safety, and adaptability. Formation control algorithms for multiple UAVs can be broadly categorized into centralized and distributed architectures. Centralized methods rely on a central processing unit for global optimization but suffer from single-point failure risks. Distributed approaches enable localized decision-making, enhancing scalability and robustness. As summarized in

Table 1. Comparison of formation control methods for multi-UAV systems.

Control Architecture	Specific Method	Advantages	Disadvantages
Centralized	Leader–Follower	Simple implementation Good formation tracking	High communication overhead Poor scalability
Distributed	Virtual Structure	High formation stability Accurate geometric control	Poor obstacle avoidance Low flexibility
	Behavior-Based	Multi-objective coordination Dynamic environment adaptation	Difficult stability analysis Empirical parameter tuning
	Artificial Potential Field	Real-time obstacle avoidance Smooth trajectory generation	Local minima problems Performance degradation in complex environments
	Consensus-Based	Dynamic topology adaptation Theoretical guarantees	Ignores agent dynamics Delay-sensitive
	Intelligent Control	Model-free operation Uncertainty tolerance	High computational load Implementation complexity

, six dominant methodologies have emerged.

Centralized methods, exemplified by the leader–follower approach, achieve global optimization through a central unit but suffer from significant communication overhead and poor scalability [5,6,7,8,9]. Among distributed methods, the virtual structure technique [10,11,12] maintains rigid formations at the cost of environmental adaptability. Behavior-based methods [13,14] enhance dynamic response through multi-behavior fusion but require empirical parameter tuning. While artificial potential field approaches [15,16,17] offer real-time obstacle avoidance that is suitable for dynamic scenarios, they frequently encounter local minima problems. Consensus algorithms [18,19,20] demonstrate strong theoretical foundations but demand high communication quality. Recent advances in intelligent control [21,22,23] incorporating machine learning techniques provide novel solutions for model uncertainty scenarios.

The leader–follower paradigm establishes explicit hierarchical relationships—where leaders generate global trajectories and followers achieve coordination through relative position/attitude tracking—demonstrating notable advantages in engineering practice [24]. Characterized by implementation simplicity, computational efficiency, and strong scalability, it has become the preferred choice for resource-constrained UAV swarms.

2.2. Blockchain

Research on the combination of drones and blockchain mainly focuses on using the decentralized, tamper-proof, and smart contract characteristics of blockchain to solve the trust management, data security and collaborative decision-making problems of drone clusters. Aujla et al. [25] proposed a blockchain-based security framework for drone-to-everything (D2X) communications to enhance the security of remote drone deployments. The authors of [26] studied the principles of blockchain and smart contracts and discussed the possibility of combining blockchain with the Internet of Things. In [27], a lightweight blockchain and a mechanism for message routing between drone nodes were proposed to improve the security of routing in 5G NR cellular networks. In [28], the authors developed a blockchain-based drone with a trusted self-organizing network mechanism (BC-UTSON) to achieve trustworthiness assessments. Zhang et al. [29] used blockchain to evaluate the trustworthiness of roadside units (RSUs) and to detect malicious RSUs in the Internet of Vehicles. In [30], the authors proposed a blockchain-based health information management framework that can access electronic medical data securely and efficiently. The author of [31] proposed the combination of blockchain and drone formation control; however, most of the research remains in the theoretical framework and fails to build a visual simulation environment. In comparison, this study achieved three-dimensional visual verification by building a high-fidelity simulation system.

Although blockchain technology has been widely used in many fields, its application research in drone formation control is still in its infancy. It is particularly noteworthy that most existing studies lack the combination of actual testing and simulation to verify the effectiveness of the model. To address this research gap, this paper proposes an innovative lightweight smart contract solution based on the distributed architecture of Ethereum that is compatible with private chains, which is easy to deploy. To verify the feasibility of the solution, this study combines blockchain technology with real drone programming tools and conducts systematic simulation experiments. The experimental results show that the solution exhibits good performance and provides an important technical foundation for the development of advanced drone systems with higher security and intelligent decision-making capabilities.

3. Model Architecture

3.1. Smart Contract

Blockchain technology uses cryptography and consensus mechanisms to enable all nodes to jointly maintain transaction records without a central authority. Each block is divided into two parts—header and body. The header contains the block version, the encrypted hash value of the previous block, Difficulty Target, Nonce, and Merkle root. The body contains the Transaction List. Each block forms an unalterable chain structure (Figure 1) through the encrypted hash value of the previous block to ensure data security.

As an extension of the blockchain, smart contracts store and automatically execute operations under preset conditions in the form of code. Based on the distributed architecture and consensus algorithm of blockchain, smart contracts allow users who do not trust each other to complete transactions without any third-party trusted intermediary or authority [32]. The operation mechanism of the smart contract is shown in Figure 2. Its core attributes include value and state, and the triggering scenarios and response rules of the contract terms are preset in the code. After the smart contract is agreed upon and signed by all parties, it is submitted to the blockchain network along with the transaction initiated by the user. After P2P propagation and miner verification, it is stored in a specific block. After the user obtains the contract address and interface information, they can enable the contract to function through transactions. Driven by the system incentive mechanism, miners contribute computing power to verify transactions and deploy or execute contract codes in the local sandbox execution environment (such as the Ethereum virtual machine) [33]. During the execution of the contract, it automatically determines whether the trigger conditions are met based on the external data source and the world state, strictly executes the preset rules, and updates the system state at the same time. The new block verified by the transaction is packaged and authenticated by the consensus algorithm, before being linked to the main chain, making all updates formally effective [34].

In this system, smart contracts serve as the core link for communication and collaboration between drone swarms. First, each drone in the swarm must be registered and verified through smart contracts, so as to effectively exclude illegal or suspicious nodes and ensure the security of the overall network. The leader submits the formation parameters or task instructions to the smart contract. The smart contract automatically executes operations and distributes key information when the preset conditions are met so that the follower can obtain the latest formation information or task instructions in real time, thereby achieving rapid formation switching and dynamic coordination on a global scale.

3.2. Leader–Follower Architecture

The leader–follower method has shown significant advantages in engineering practice by defining a clear hierarchical relationship, whereby the leader generates the global trajectory, and the follower achieves collaboration through relative position or attitude tracking [24]. With its simple implementation, high computational efficiency, and strong scalability (supporting the dynamic addition and subtraction of nodes), it has become an ideal choice for resource-constrained drone clusters.

The formation control in this paper adopts an improved pilot–follower architecture. The pilot node serves as the cluster admission controller, and the identity registration and permission management of the follower drone nodes are achieved through the blockchain. Unauthenticated nodes will be denied access to the blockchain network and cannot enable any smart contract functions. As shown in Figure 3, the pilot registers the drone to the blockchain through the smart contract interface. Authorized drones can access the blockchain and obtain target data. When unauthorized illegal drone nodes access the blockchain, they will be denied access, which effectively prevents network attacks by malicious nodes.

Figure 4 describes the process of issuing instructions. After registration, the pilot can write the formation control parameters (including the pilot coordinates and the formation task type) into the blockchain. The follower obtains the formation parameters by reading the on-chain data and selects the appropriate position according to different formation types. The specific selection process is described in Section 4.2.

4. Methods

4.1. Smart Contract

As a core component in the blockchain system, smart contracts are used to ensure the secure communication and dynamic formation control of drone swarms. In our system, smart contracts are written in the Solidity programming language and are deployed on a blockchain network compatible with the Ethereum Virtual Machine (EVM).

Solidity is a high-level language designed specifically for smart contracts that supports object-oriented development and is suitable for building highly reliable and automatically executed contract logic [35]. Contracts can accurately control the registration process of drones, the creation and management of tasks, the dynamic location allocation of each node in the formation, and the data chaining and verification operations during task execution. In addition, Solidity’s powerful event mechanism and mapping structure make data storage and state management more efficient and transparent. Combining the advantages of blockchain, such as decentralization, immutability, and full traceability, the smart contract we designed achieves a secure and reliable scheduling and communication mechanism for drone swarms when performing multi-task and multi-role collaboration.

First, the contract defines the enumeration type formation, which is used to describe the task type and the corresponding formation form (such as Line, V, and Circle). This design makes the drone group highly flexible in task scheduling and formation control, and can adjust the cooperation mode of the drone group according to different scenarios. For example, when the formation type is linear, it is suitable for tasks such as crossing narrow terrain or channels, uniform coverage, and efficient scanning. The name of the task can be pesticide spraying; when the formation type is circular, it can achieve 360° all-round encirclement of the target object, and the name of the task can be defensive encirclement. When the formation type is V-shaped, it can achieve long-distance cruising missions by reducing the energy consumption of the rear drone, and the name of the task can be border patrol. In addition, the contract achieves the management of drone registration, location allocation, and task information through mapping.

To ensure the orderly management and coordinated flight of drone clusters, smart contracts provide a mechanism for registering/deleting identities. The leader in the cluster has the sole authority to add or remove drones (RegisterDrone and RemoveDrone functions). Smart contracts support the creation and modification of missions by the leader and the real-time acquisition of missions by the follower (CreatMission, UpdateMission, and GetMission functions), ensuring that missions can be dynamically adjusted according to the scenario. To achieve trusted data sharing, smart contracts allow each registered drone to upload its collected environmental data, coordinate location, and other information to the blockchain for other drones to query and analyze missions (SubmitData and GetData functions).

Driven by the incentive mechanism, miners verify the transaction through the P2P network and consensus algorithm, and the contract status will be updated and written into the new block to ensure that all operations are tamper-proof, open, and transparent [36]. The design and implementation of the entire smart contract enables efficient collaboration and secure communication among drone groups in highly dynamic environments through automated execution and state management.

4.2. Dynamic Formation Control

In Ethereum, a “node” usually refers to an instance that runs an Ethereum client and maintains the state of the blockchain, while an “account” refers to a subject that has a unique address and private key on the blockchain [37]. In the Ganache simulation environment, we can regard each account as a “drone” with an independent identity, and use their addresses to distinguish different individuals, thereby meeting the need for “multiple entity interactions” in the simulation scenario.

The core mechanism of the leader includes the following two levels: At the member management level, the leader achieves the dynamic registration and deregistration of the follower through the registerDrone and removeDrone interface functions provided by the smart contract. All drone address information is stored through the mapping structure mapping (address => bool) DroneRegistry, where the Boolean value indicates the registration status. At the parameter management level, the leader submits the formation control parameters (mission type and location coordinates) to the specific data structure of the blockchain through the CreateMission and SubmitData interface functions; at the same time, the leader can change the formation type at any time and re-upload new data to achieve dynamic formation control. The core mechanism of following the randomness is reflected in the data verification level. The built-in verification logic of the smart contract ensures that all operations must be authenticated by digital signatures, and the data cannot be tampered with through the consensus mechanism. Followers can now check the mission and declared formation by reading the data and mission submitted by their leader. Knowing the position of the leader, they can choose the formation position that is closest to them.The operation flow chart of the leader drone and the follower drone is shown in Figure 5.

To enable flexible support for multiple formation configurations, we proposes a formation control method based on dynamic adjustment of angle and distance parameters. The method introduces two key set variables—formation_distances (representing the set of predefined distances between the leader and follower UAVs, denoted as $f_d$) and formation_angles (representing the set of predefined relative angles of followers with respect to the leader, denoted as $f_a$). The relative angle $\theta$ and base distance d can be customized by users according to the actual application scenarios.

When executing formation control logic, each follower first obtains the current number of followers n in the formation from the blockchain. Based on the command type issued by the leader, the total number of nodes n, as well as parameters $\theta$ and d, the system dynamically adjusts the values of sets $f_d$ and $f_a$. Subsequently, each follower completes coordinate calculation, relying on these sets to achieve flexible position distribution control, thereby granting the UAV formation excellent scalability. The following sections will introduce specific control logic implementations by examining three typical formation configurations.

In the V formation, the angle set is defined as $f_a=\{\theta ,-\theta \}$, and the distance set is defined as $f_d=\{d,2d,\dots ,kd\}$ (where $k=n/2$). Followers are symmetrically distributed behind the leader at different distances ($d,2d,\dots ,kd$) with angles $\theta$ and $-\theta$. The parameters $\theta$ and the base distance d can be predefined by the user. As shown in Figure 6, the V-formation configurations are illustrated for $\theta ={30}^{\circ }$, $n=4$ and $\theta ={45}^{\circ }$, $n=6$.

In line formation, the angle set is defined as $f_a=\{{90}^{\circ },-{90}^{\circ }\}$, and the distance set is defined as $f_d=\{d,2d,\dots ,kd\}$ (where $k=n/2$). Followers are distributed symmetrically on both sides of the leader at different distances, perpendicular to the leader’s heading (at ${90}^{\circ }$ and $-{90}^{\circ }$). As shown in Figure 7, the line formation configurations are illustrated for $n=4$ and $n=6$.

In circle formation, the equal division angle $\beta$ is first calculated based on the total number of followers: $\beta ={360}^{\circ }/n$. The angle set is defined as $f_a=\{k\beta \mid k=1,2,\dots ,n\}$, and the distance set is defined as $f_d=\left\{d\right\}$. Followers are uniformly distributed around the leader at the same distance d with angular spacing $\beta$, forming a closed circle. As shown in Figure 8, the circle formation configurations are illustrated for $n=4$ and $n=6$.

To determine the follower’s geographic coordinates from the relative bearing angle $\theta$ and distance d to the leader, a polar-to-Cartesian coordinate transformation is required. For large-scale formations where Earth’s curvature cannot be neglected, the Haversine formula provides a spherical geometry-based solution. If the coordinates of two points are $({\varphi }_1,{\lambda }_1)$ and $({\varphi }_2,{\lambda }_2)$ (latitude and longitude), the distance between the two points can be determined using the following equation:

$$cos\delta =sin{\varphi }_1·sin{\varphi }_2+cos{\varphi }_1·cos{\varphi }_2·cos({\lambda }_2-{\lambda }_1)$$

(1)

$$\mathrm{distance}=r·\delta$$

(2)

where $\Delta \varphi ={\varphi }_2-{\varphi }_1$, $\Delta \lambda ={\lambda }_2-{\lambda }_1$, and r is the radius of the sphere (e.g., the average radius of the Earth is approximately 6371 km). $\delta$ is the central angle between two points. Then, when the central angle $\delta$ and the distance are known, the target position $({\varphi }_2,{\lambda }_2)$ can be inferred by the following formula:

$${\varphi }_2=arcsin\left(sin{\varphi }_1·cos\left({\displaystyle \frac{d}{r}}\right)+cos{\varphi }_1·sin\left({\displaystyle \frac{d}{r}}\right)·cos\delta \right)$$

(3)

$${\lambda }_2={\lambda }_1+arctan2\left(sin\alpha ·sin\left({\displaystyle \frac{d}{r}}\right)·cos{\varphi }_1,cos\left({\displaystyle \frac{d}{r}}\right)-sin{\varphi }_1·sin{\varphi }_2\right)$$

(4)

To generate all possible candidate positions of followers for a specific formation type, we employ a candidate position generation algorithm that iterates through the angle set $f_a$ and distance set $f_d$. For each $(d,\alpha )$ combination, spherical coordinate transformation is performed using Equations (3) and (4) to compute the corresponding follower coordinates. This process ultimately produces a leader-referenced candidate position set, which serves as the foundation for subsequent position screening and selection. The complete procedure is shown in Algorithm 1.

Algorithm 1 Generate candidate positions

Require:  Leader position $({\varphi }_1,{\lambda }_1)$, distance set $f_d$, angle set $f_a$ Ensure:  Set of candidate positions C   1: $C\leftarrow \varnothing$   2: for all  $d\in f_d$ do   3:     for all $\alpha \in f_a$ do   4:         $({\varphi }_2,{\lambda }_2)\leftarrow \mathrm{CALCULATE}\text{\_}\mathrm{FOLLOWER}\text{\_}\mathrm{COORDINATES}({\varphi }_1,{\lambda }_1,d,\alpha )$    ▹ Use Formulas (3) and (4) to perform spherical coordinate transformation   5:         $C\leftarrow C\cup \left\{({\varphi }_2,{\lambda }_2)\right\}$   6:     end for   7: end for   8: return C

After obtaining all the target positions, the following Euclidean distance formula can be used:

$$r_i=\sqrt{{({\lambda }_d-{\lambda }_i)}^2+{({\varphi }_d-{\varphi }_i)}^2}$$

(5)

Then, the distance between all target positions and itself can be calculated in order to select the closest target position.

Taking Figure 9 as an example, in a V-shaped formation, the leader is fixed at position 0 as the formation reference point, and the follower is initially at a random position. The follower will calculate positions 1, 2, 3, and 4 based on the leader’s position and Formulas (1) and (2), obtaining their Euclidean distances r from themselves. It will then select the closest position—position 1 in this example. However, the follower will not fly to the position immediately, but will first confirm whether position 1 is free through the blockchain; if it is free, it will fly directly to the position. If the position is occupied by other drones, it will select the next closest position (such as position 3). Once the specific position is determined, the follower will record the selection in the blockchain and mark the position as “occupied”, thereby effectively avoiding the risk of spatial position conflicts among multiple drones and achieving a reliable reconstruction of the formation system.

4.3. Simulation Environment

Our simulation experiment is built on a virtual machine environment based on the Ubuntu 20.04 LTS operating system (64-bit, X64 architecture). The virtual machine is deployed on the physical host through the VMware virtualization platform, and a 4-core CPU and 8 GB memory are allocated to support multi-task parallel computing requirements. The hardware configuration of the physical host is an AMD Ryzen 7 4800H processor and 16 GB DDR4 memory.

The core components of the simulation environment include the following modules:

Blockchain network simulation: Ganache v2.13.2 was used to build a private blockchain test network. Ten initial test accounts were configured—the network ID was 5777 and the RPC port number was 8545, which was used to simulate node communication and smart contract deployment in a distributed ledger environment.

Drone dynamics simulation: The drone model is built based on PX4 Autopilot v1.13.0 firmware, and the Gazebo 11.0 simulation engine is integrated to simulate the dynamic characteristics and sensor data (such as IMU and GPS) of multi-rotor drones. The experimental scene is achieved through Gazebo’s 3D physics engine, which supports dynamic obstacles and multi-machine collaborative flight testing.

Communication and control framework: The communication between the drone and the ground control station (GCS) is achieved through the MAVSDK-Python v1.4.4 library, and a real-time data link is established using UDP port 14540. At the same time, the Web3.py v6.0.0 library is integrated to encrypt the drone status data (such as position, speed, and mission log) and store it in the blockchain network to ensure data immutability and traceability.

Multi-node collaborative testing: NAT network mode is configured in the virtual machine to ensure low-latency communication between Gazebo, PX4, and Ganache nodes. Blockchain transaction verification and the synchronization of drone control instructions are implemented through custom Python scripts. The test scripts are developed based on Python 3.8 and rely on ROS Noetic to optimize task scheduling.

5. Results

We built a drone swarm consisting of one leader drone and four follower drones, and evaluated the model’s functionality by implementing accurate simulations in real flight scenarios based on the Gazebo high-fidelity physical simulation environment and the MAVSDK framework of the MAVLink communication protocol.

Figure 10 and Figure 11 illustrate the multi-UAV formation control process with four followers and eight followers, respectively. During the system initialization phase, both the leader and follower drones remain in standby mode. When the leader writes the formation task information into the blockchain network via the smart contract, each follower node can monitor the instructions in real time through the on-chain event-triggered mechanism and can execute corresponding cooperative control operations accordingly. The experiment validated three typical formation types—V-shaped formation, line formation, and circular formation. The results demonstrate that the UAV swarm can smoothly switch between different formation types according to mission requirements, exhibiting excellent scalability in relation to the number of UAVs.

In conventional leader–follower architectures, follower UAVs rely entirely on centralized command control from the leader aircraft. As formation size increases, this architecture faces challenges of exponentially growing control complexity and communication overhead. Our study introduces an innovative blockchain-based event-driven mechanism, whereby follower UAVs autonomously acquire mission parameters and complete formation adjustments by monitoring on-chain events, achieving decentralized cooperative control. To evaluate real-time system performance, as shown in

Table 2. Response latency under different fleet sizes.

Formation Change	4-UAV Latency (s)	8-UAV Latency (s)
1	1.68	1.74
2	1.70	1.76
3	1.72	1.79
4	1.72	1.79
5	1.70	1.76
6	1.71	1.77
7	1.70	1.79
8	1.72	1.80
9	1.65	1.70
10	1.70	1.80
Max Latency	1.72	1.80

, we recorded the response latency of the last UAV in 10 formation-switching scenarios. By comparing latency data between 4-UAV and 8-UAV configurations, we can clearly analyze performance trends during system scaling. The Ganache framework we employ utilizes an instant mining mechanism that guarantees block generation within milliseconds, effectively eliminating the stochastic delays inherent in Proof-of-Work (PoW) consensus. Consequently, our analysis focuses exclusively on the system’s response latency.

The experimental results show that under the blockchain event-driven mechanism, when the number of follower drones increases from 4 to 8, the maximum response delay only increases from 1.72 s to 1.80 s—an increase of less than 5%—exhibiting a slight upward trend. This increase is primarily attributed to the higher complexity of formation position computation and the increased system load. However, due to the parallel nature of event broadcasting, the overall control latency remains low. In contrast, related studies have shown that the delay in centralized control systems typically increases by more than 20% under similar conditions. These results demonstrate that the blockchain event-driven control mechanism offers superior scalability and robustness in large-scale swarm formation scenarios.

In the Ganache-provided local Ethereum simulation environment, we map each account to a UAV node and implement task allocation and position management through deployed smart contracts. As shown in Figure 12, each block in Ganache stores transaction hashes along with detailed transaction values, ensuring the integrity of all data. Furthermore, the chained structure of blocks provides cryptographic guarantees of data immutability.

6. Conclusions

Our research addresses the challenges in dynamic formation control and secure communication for multi-UAV swarms by proposing a novel blockchain-based distributed control architecture. Leveraging the decentralized nature, tamper-proof characteristics, and automated execution capabilities of blockchain smart contracts, this architecture achieves secure UAV identity authentication, task scheduling, and real-time data sharing, effectively overcoming the limitations of traditional centralized approaches in terms of security, robustness, and response speed.

Through the implementation of a local blockchain simulation environment using Ganache and a high-fidelity UAV dynamics simulation platform based on Gazebo, we experimentally validated the system’s capability for dynamic switching among V-shaped, linear, and circular formations. The results demonstrate that under the blockchain event-triggered mechanism, the UAV swarm maintains low-latency response and position adjustment even as follower nodes increase, ensuring efficient coordination while significantly reducing risks caused by communication delays and scheduling errors. Furthermore, the blockchain-based transaction recording guarantees the transparency and immutability of mission data, providing robust security assurance for UAV swarm operations in complex environments.

However, this study has certain limitations and areas for improvement. First, the current validation relies primarily on simulation environments without real-flight testing, thus requiring further investigation of practical factors like communication latency, channel interference, and environmental noise. Second, while we implemented three typical formation types, dynamic switching between multiple formation patterns could be enhanced through algorithmic extensions. Additionally, our Ganache-based private Ethereum chain employs instant mining, neglecting PoW-induced delays; in practical applications, lightweight consensus algorithms could be adopted to minimize mining latency and improve real-time control performance.

In conclusion, this research represents meaningful exploration and innovation in blockchain-driven dynamic formation control for multi-UAV systems, offering new technical perspectives for secure and intelligent swarm coordination. Future work will focus on optimizing system modules to enhance operational adaptability and scalability, ultimately delivering more reliable and efficient solutions for military reconnaissance, emergency response, logistics delivery, and related fields.