A robot swarm consists of a large number of autonomous robots that exploit self-organization to coordinate their activities and to solve complex problems that are beyond the capabilities of the single robots. Even though the inherent robustness to failure, scalability, and flexibility of a self-organized system makes of robot swarms ideal candidates for a number of real world tasks, current studies have limited their attention to safe laboratory settings and have virtually ignored security issues.
The main goal of my PhD project is to build a secure robot swarm that is scalable to real-world applications. To enable this, I will address security issues such as how to certify the swarm behavior and how to make the swarm behavior tamper proof in a fully decentralized way. In the project, the swarm members will interact and coordinate via a blockchain, a distributed database first developed in the context of the digital currency Bitcoin. The blockchain is intended to establish a secure communication medium and financial reward based coordination system for the robot swarm. Smart contracts—programming code on the blockchain that is automatically executed if a specified event occurs—will introduce unstoppable and secure coordination mechanisms using the Ethereum framework. The project seeks to investigate possibilities and limitations of combining blockchain and swarm robotics technologies to produce secure robot swarms.
I will conduct experiments to test robot swarm coordination mechanisms based on smart contracts and blockchains and I will analyze the obtained swarm dynamics and evaluate their robustness in a number of different scenarios. I will study how financial rewards can help to control the robot swarm complex dynamics and to limit/remove the potential damage of rogue robots. My objectives are to identify critical steps in designing secure systems, to show implications of attacker strategies, and to propose solutions to security challenges in swarm robotics.