In this paper, a Distributed Nonlinear Dynamic Inversion (DNDI)-based consensus protocol is designed to achieve the bipartite consensus of nonlinear agents over a signed graph. DNDI inherits the advantage of nonlinear dynamic inversion theory, and the application to the bipartite problem is a new idea. Moreover, communication noise is considered to make the scenario more realistic. The convergence study provides a solid theoretical base, and a realistic simulation study shows the effectiveness of the proposed protocol.
Keywords: Distributed Nonlinear Dynamic Inversion; bipartite consensus; communication noise; signed graph.