Zero-Trust-Based Privacy-Preserving Distributed Localization for Mobile Robot Networks

Abstract

Localization is the foundation for achieving autonomous navigation of robots, and protecting the privacy of robots during the localization process is crucial. This paper develops a zero-trust-based privacy-preserving distributed iteration localization (ZTPP-DILOC) algorithm for a robot network that moves freely within bounded areas. In this algorithm, based on the principle of “never trust, always verify” in the zero-trust security framework, real-time trust evaluation mechanisms are established for both sent and received information. On the one hand, each robot evaluates its trust degree for nearby robots, and integrates the historical information to generate false location estimation values for sending, thereby protecting the security of information transmission. On the other hand, each robot evaluates its trust degree for the received information, and generates a time-varying iterative equation, thereby protecting the security of location estimation update rules. A comprehensive analysis is conducted on the convergence, complexity, and privacy-preserving performance of the ZTPP-DILOC algorithm, relying on edge combination and sub-stochastic matrix techniques. The ZTPP-DILOC algorithm is shown to be capable of accurate localization without disclosing the real locations of robots. At last, the effectiveness of the algorithm is verified through computer simulations and experiments.