A novel LPV sector approach for nonlinear dynamic inversion consensus control of multi-agent systems under saturating input nonlinearities

This paper investigates the consensus control of linear multi-agent systems (MASs), facing a wide class of input saturating or sector nonlinearities, by application of the nonlinearity transformation and dynamic inversion. A fundamental linear parameter-varying (LPV) theory-based regional sector condition has been derived to model the nonlinearity of actuator saturation. The proposed model considers a local sector to develop a regional model, and this approach is more general than existing methods. The suggested sector condition is applied to the consensus control issue for linear leader-following MASs in the presence of input saturation. A new nonlinear control, by application of nonlinear dynamic inversion (NDI) approach, is proposed to deal with the leader-following consensus problem of multiple systems with the proposed LPV-based sector condition. The local leader-following consensus control condition with an assured domain of attraction has been provided by applying Lyapunov stability, NDI method, regional sector condition, LPV theory, and algebraic transformations under saturation and directed topology. In comparison with traditional techniques, the proposed method applies to a wide range of saturation nonlinearities, guarantees an extensive stability region, and addresses both symmetric and asymmetric saturation functions. The novel outcomes achieved through the proposed methodology, as specific scenarios, involve system stabilization and the achievement of consensus control among followers within an undirected topology. The linear consensus under input saturation results are also extended for the Lipschitz nonlinear MASs with a guaranteed region of stability. The simulation results of the proposed methodology for consensus control, accompanied by a comparison with existing techniques, indicate the superiority of the resulting approach over existing methods.