Finite-time bipartite synchronization of multi-agent systems is assessed here. To cover a wide range of practical systems, the agents dynamics are modeled as nonlinear nonstrict-feedback dynamics with unknown time-varying control coefficients. Moreover, to conform to many real-world applications, the agents dynamics are assumed unknown and their state information is assumed not available for measurement. First, a virtual affine variable is introduced, and to eliminate the algebraic-loop in control design, neural network along with minimal learning parameter principle are employed to approximate composite uncertainties including unknown functions in the system dynamics, unknown control coefficients and control inputs. A local linear state-observer is then designed for each follower agent to estimate unmeasured states, and finally dynamic surface control scheme and finite-time approach are adopted in the distributed bipartite control design to guarantee finite convergence time. It is ensured that under this proposed protocol, the bipartite error surfaces and boundary error layers are converged to a neighborhood of the origin in the sense of finite-time stability. Finally, two simulation examples clarify the efficiency of the designed controller.

Keywords