A Bio-Inspired Filtering for Multi-Robots Localization via Stochastic Gene Regulatory Network

Le résumé fourni par la source

Abstract In multi‑robot environments, achieving precise localization and trajectory tracking is complicated by incomplete system modeling and fluctuating measurement noise. This work introduces a bio‑inspired proprioceptive filtering framework grounded in a stochastic gene regulatory network (SGRN) model. The method estimates both static and dynamic states through relative state evaluation and feature‑based matching. A multi‑objective optimization process is employed to fine‑tune the SGRN filter, simultaneously minimizing absolute error and settling time while maintaining computational feasibility within the constraints of robot dynamics. Simulation studies demonstrate notable reductions in localization error and improved swarm coordination, confirming the robustness and scalability of the proposed approach.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.