Secure Multi-Robot Adaptive Information Sampling

Abstract

In a coordinated multi-robot information sampling scenario, robots often share their collected information with others for a better prediction. As with any other online data sharing technique, data integrity is a concern, but it has not yet been addressed in the multi-robot information sampling literature. In this paper, we study how to secure the information being shared among the robots in such a multi-robot network against integrity attacks and what is the cost of integrating such security techniques. To this end, we propose a Blockchain-based information sharing protocol that helps the robots reject fake data injection by a malicious entity. On the other hand, optimal information sampling is a compute-intensive technique and so are the popular Blockchain-based consensus protocols. Therefore, we also study the impact of adding such a security protocol on the execution time of the sampling algorithm, which in turn effects the energy spent by the robots. Results show that our proposed technique is effective against such data tampering attempts while the effect of the added computation varies largely on the consensus protocol used.