Blockchain approach to solve collective decision making problems for swarm robotics

Abstract

Recently, there are significantly increasing number of applications of swarm robotics for example targeted material delivery, precision farming, surveillance, defense and many other areas. Some qualities such as robot autonomy, decentralized control, collective decision-making ability, high fault tolerance etc. make swarm robotics suitable for solving these real-world problems. Blockchain, a decentralized ledger which is managed by a peer-to-peer network with cryptographic algorithms, provides a platform to perform different transactions in a secure way. The decentralized nature of swarm robotics makes it compatible to combine with blockchain technology and allows it to implement different decentralized decision making, behavior differentiation and other business models. This paper proposed a new distributed collective decision-making algorithm (best-of-n) for swarm robotics, where robots form a peer-to-peer network and perform different transactions using blockchain technology. For performance comparison, both collective decision-making algorithm with and without using blockchain have been implemented and their results have also been compared with respect to different metrics such as consensus time and exit probability. The performance shows that our proposed method gives a lower consensus time and higher exit probability value and therefore outperforms classical methods.