Abstract:Swarming is a disruptive, game-changing technology that involves the coordinated deployment of multiple unmanned systems in multiple domains (land, sea, air, space) . Developments in new microelectronics, guidance, navigation, sensors and artificial intelligence technologies enable low-cost micro-drones to perform challenging missions. When combined with novel decision-making processes, target tracking, communications technologies and algorithms, swarms can have a huge impact on the battlefield. It can also be used to provide continuous and undetectable surveillance capabilities as well as critical defense capabilities, such as intercepting cruise missiles. This article describes European research activities in the field of swarming and delves into the significant implications it may have for defence.
Keywords: UAV, autonomous capabilities, target tracking, guidance and control, sensor fusion
Drone swarms are composed of a variety of different unmanned equipment. They have the capabilities of decision-making, target tracking, guidance and control, sensor fusion and command, and can operate as a group of intelligent autonomous "systems". Each capability in the swarm system can also operate independently as an independent intelligent autonomous system
Design unmanned vehicles with the necessary intelligence and autonomy by leveraging developments in unmanned devices, artificial intelligence (AI), communications, guidance and control, sensor fusion, aerospace and unmanned traffic management (UTM) These drones can fly in a coordinated manner like a swarm of bees or a flock of birds. Such drone swarms could bring game-changing capabilities to defense companies that combine digital and smart technologies. Drone swarms can also help Europe enhance security capabilities and have a disruptive impact on other areas such as urban transportation, unmanned traffic management, autonomous driving and unmanned equipment
In terms of national defense, the use of drones can significantly reduce the risk of combatants being exposed to dangerous environments. Drones are capable of performing boring and laborious tasks such as continuous monitoring of large areas (e.g. the Mediterranean Sea, European borders). In future military scenarios, drones can also be used to confuse and overwhelm opponents. Figure 1 illustrates the multi-domain drone swarm concept for the protection and surveillance of high-value assets such as military camps or installations, developed in 2016 under the European Defense Agency pilot project "European Swarm" (EUROSWARM) first proposed.
Figure 1: Multi-domain drone swarm concept
A drone swarm consists of a group of autonomous or semi-autonomous drones. The drones in the swarm can cooperate with each other to achieve a common goal. Drone swarms can be used for a variety of military missions such as reconnaissance, surveillance and strike operations. When it comes to drone swarm design, military applications must consider the following four main technology modules. From sensing capabilities to swarm communication protocols and routing, each module is important to drone swarm performance and stability.
1. Perception capability: Perception capability is crucial for drone swarms. This capability enables the swarm to sense and understand its environment, detect obstacles, identify targets, and maintain situational awareness. Effective sensing capabilities enable bee colonies to work in complex dynamic environments and perform various complex tasks efficiently. Machine learning and artificial intelligence technologies can enhance perception capabilities. Swarms can achieve unprecedented situational awareness through sensor fusion.
2. Task allocation and decision-making: Task allocation and independent decision-making are crucial to drone swarms. Efficient tasking ensures that each drone is assigned a mission that matches its capabilities, optimizing the use of available resources. If one or more drones fail, the swarm can automatically fill the gap, and task allocation helps the rapid dissemination of decisions, allowing the swarm to quickly adapt to changing environments. Task distribution also helps improve adaptability, scalability, and speeds up decision-making, making swarms more effective in dynamic and uncertain environments. Through seamless data exchange between drones within the swarm, the swarm is able to make better decisions, which improves efficient resource utilization, stability, fault tolerance, adaptability and scalability.
3. Path planning and collision avoidance: Drone swarms usually consist of a large number of drones. In order to achieve efficient and safe operation of drone swarms, path planning and collision avoidance methods are crucial. The core of swarm path planning is to find the best path for each drone to reach its destination while avoiding obstacles and minimizing the time and energy required. Collision avoidance ensures that drones will not collide, allowing each drone to successfully complete its mission. For example, in surveillance missions, path planning can optimize the drone's route of action to minimize overlapping areas and increase coverage. Currently, there are a variety of techniques to achieve path planning and collision avoidance, including centralized methods and decentralized methods. The centralized approach involves the planning of individual drones and the coordination of the paths of all drones in the swarm. The decentralized approach involves each drone making its own path planning decisions based on local information.
4. Communication: A swarm operates optimally when communication between drones within the swarm remains open and without delays. Through sensor fusion, swarms can provide high-determinism and high-resolution information. Effective communication protocols enable drones to share information such as location, status, and task assignments, while routing is responsible for finding the best path for information dissemination between drones. This enables swarms to work together, coordinate actions, and share information in real time. Currently, there are various technologies that can implement communication protocols and routing, such as ad hoc network technology, mesh network technology, and multi-hop routing technology. Ad hoc networks enable drones to communicate directly with each other without the need for fixed infrastructure. Mesh networks, on the other hand, use drones to form a network with redundant paths for communication. Multi-hop routing can realize relay transmission of information between drones, thereby achieving longer-distance communication.
The communication protocol of UAV swarms can be adjusted according to mission requirements and swarm characteristics to achieve information exchange. Currently, there are three main architectural approaches to designing military drone swarms, which include:
1. Centralized architecture: In this approach, the operations of all drones within the swarm are coordinated by a central control entity, such as a ground control station. The central control entity is able to communicate with the swarm and collect data, process the data, and make decisions. This method is suitable for small colonies and simple tasks.
2. Decentralized architecture: In this approach, there is no central control entity, and each drone in the swarm is able to operate independently, make decisions based on local information, and interact with Other drones carry out information communication. This approach is suitable for large-scale swarming missions and other highly complex tasks.
3. Hybrid architecture: This approach combines the advantages of centralized and decentralized architectures. In this approach, there is a central control entity that provides high-level navigation to the drones, while each drone is equipped with autonomous decision-making capabilities.
Drone swarms have a wide range of applications in the military field and can perform a variety of tasks. The following are key examples of some types of autonomous swarm operations:
1. Area coverage: In area coverage operations, the swarm’s task is to scan a specific area using sensors equipped with drones. In most cases, the ideal area coverage is 100% and the drone must completely scan the area. Covering an area with multiple drones can encounter some problems, and common ways to deal with these problems include using decomposition techniques to divide the focus area into a set of sub-areas and deploying a drone in each sub-area. For swarms containing different types of drones, the range of the sensor, the maneuverability of the drones, and the range must be considered during the area decomposition process to improve the efficiency of the system. After each drone is assigned a sub-area, these drones need to independently plan their paths within their respective areas. Coverage path planning methods include 2D, 3D and multi-UAV area coverage.
2. Comprehensive and sustained area coverage: Comprehensive and sustained area coverage requires the deployment of drone swarms and their ability to provide sensor coverage throughout a given area throughout the mission. The drones within the swarm should form a formation based on the location of their sensors and possible environmental features such as obstacles or occlusion areas. UAVs within the swarm are deployed statically or dynamically as regional characteristics or monitoring areas evolve over time. The main purpose is to design a formation pattern that achieves full static coverage with a minimum number of drones.
3. Area search: In area search operations, the task of the drone swarm is usually to search for specific targets in key areas. In this operation, complete coverage of the area is not required. The swarm must identify targets in the area in the shortest amount of time. The drones in the swarm must cooperate throughout the mission, and they will use online decision-making and path planning techniques to improve the performance of the system based on their own perception of the environment and the behavior of other drones. A swarm area search algorithm can be used to predict the probability of a target distribution. Bionic swarm algorithms have also attracted scientific interest in area search operations.
4. Area surveillance: The area surveillance operation requires the swarm to continuously monitor a specific area. Area surveillance is typically used for patrolling, surveillance, detection of emergent or dynamic threats, and border security. This action is intended to minimize the time between two long monitoring periods.
5. Target tracking: Ordinary target tracking operations involve a target and a drone. The drone's tracking range is based on online path planning based on its sensor data and an estimate of the target's location, and in some cases the target's predicted behavior or future location. The drone must navigate on its own so that it always follows its target. With the introduction of swarm capabilities, target tracking tasks can be completed by multiple drones that can track single or multiple targets.
The design of military drone swarms requires comprehensive consideration of multiple factors, including mission requirements, swarm size, communication capabilities and computing resources
1. Low observability unmanned aerial vehicle system - "LOTUS"
The "Lotus" project is a European Defense Industrial Development Program (EDIDP) project consisting of 9 companies from Greece and Cyprus and another 2 companies from Spain and the Netherlands. The project, led by lntracom Defense, was launched in December 2020 and will last for 45 months. The Lotus project team has designed an advanced unmanned aerial system for tactical aerial reconnaissance and surveillance missions. The UAV possesses several key features, including stealth capabilities, stand-off combat capability, airworthiness, interoperability based on NATO standards, and reliable communications taking into account network security. In addition, the system extensively uses artificial intelligence technology to ensure that it can accurately perform complex tasks. The UAV mothership designed by the project team is equipped with multiple infrared sensors, has low observability and long endurance, and is also equipped with a self-defense system to deal with enemy threats. The UAV carrier can deploy four tube-launched folding-wing UAVs with advanced autonomous capabilities and the ability to perform complex ISR missions. The drone mothership and drones together form a powerful drone swarm, which can achieve delay-free collaboration and provide critical intelligence and monitoring data to on-site decision-makers. In Figure 2, intelligence tasking (left) and collaborative coverage (right) of ground targets are performed via a swarm algorithm developed at the University of Patras.
Figure 2: Example of drone swarm action in the "Lotus" project
2. Autonomous, reconfigurable drone swarms for defense applications - "ACHILLES". Rewritten content: ACHILLES is an autonomous, reconfigurable drone swarm for defense applications
The "Achilles" project is a European Defense Agency project, a collaboration between Greek and German companies and universities. The project was launched in January 2023 and is led by the University of Patras. Companies and schools participating in the project include ATOS, DroniQ, Scytalvs, Intracom Defense, University of Patras, Ingolstadt University of Applied Sciences and the University of Athens. This project aims to promote the development and use of drone swarms in the defense sector by increasing the TRL of autonomous, reconfigurable drone swarms for specific defense missions, and to demonstrate drone swarms for continuous surveillance capabilities and combat readiness levels. The project team discovered a variety of advantages and potential applications of drone swarms, and used this as inspiration to promote the Achilles project. Recent technological advances enable drones to automatically acquire critical data to enhance situational awareness. Scalable autonomous and reconfigurable drone swarms have highly effective regional coordination capabilities and strong adaptability to various events. Outcomes and innovations from the project are expected to include new drone swarming capabilities and methods to enable the safe and effective integration of drones into military and civilian airspace. These capabilities will support the maturation and validation of drone swarm-based systems and technologies.
3. Manned-unmanned system escort operations - "COMMANDS" project Rewritten content: 3. Manned-unmanned system escort operations - "COMMANDS"
The "Command" project is a European Defense Fund (EDF) project involving 21 companies from 10 countries. Launched in December 2022, the three-year project is led by Sener Aerospace and is supported by the defense ministries of seven countries. The ministries of defense of these countries also provide the project with basic requirements for project development. The goal of the project is to develop agile, intelligent and collaborative manned and unmanned system life cycle capabilities (TLC). The project will include several modular systems that enable swarming capabilities through seamless functional services and data exchange. The system will consist of manned and unmanned ground vehicles and drones. To reduce project risks, the project uses a variety of technologies to develop a self-sustainable EU Defense TLC sustainability roadmap. Ultimately, the project will upgrade existing ground vehicles and be integrated into future vehicles. The project plans to validate the technology through technology demonstrators, which will be conducted in laboratories and mobile demonstrators in real-life scenarios, with a focus on providing armed escorts for convoys performing "last mile" resupply
Swarm technology enables large numbers of drones to be highly interconnected, effectively plan and assign mission objectives, make coordinated tactical decisions, and collaboratively respond to dynamic environments with minimal supervision while making recommendations to operators. As swarm technology matures, its applications in the military field are also developing. Many believe that the development of swarms can be compared to the development of precision weaponry. Precision-equipped weapons were tested and refined in the 1970s and 1980s, but did not emerge until the first Gulf War in the early 1990s. The use of swarms could render obsolete manned defense systems currently used for low- and medium-altitude surveillance and strike missions. In the coming decades, remotely operated single unmanned equipment such as drones will become obsolete as unmanned aerial systems in the air, on the ground and at sea will be able to deploy multiple drones to fight in swarms, This expands the combat range and surveillance range, and improves the UAV's ISTAR capabilities and attack capabilities for performing various missions.
NATO’s multinational defense ministries have announced their intention to integrate swarm technology with existing weapon systems, such as F-35 fighter jets, British Storm next-generation fighter jets, and FCAS aircraft/unmanned aerial systems. It’s clear that swarm technology is being integrated with military capabilities around the world and starting to impact countries’ defense capabilities. Swarms are a key military technology that will directly facilitate the development of multiple technological fields. For example, this technology can optimize the artificial intelligence technology of embedded swarms and autonomous systems, effectively improving the safety, efficiency and effectiveness of collaborative operations of defense systems operating in unstructured, rapidly changing, restricted and confrontational environments. As seen in the Russia-Ukraine war and conflicts in the Middle East, swarm technology is changing warfare through the use of swarms of drones and loitering munitions. The development of autonomous swarm systems is therefore vital to Europe’s defence, security and prosperity.
The above is the detailed content of Drone swarms: A game-changing technology for defense applications. For more information, please follow other related articles on the PHP Chinese website!
![[Web front-end] Node.js quick start](https://img.php.cn/upload/course/000/000/067/662b5d34ba7c0227.png)
