FFDL-based sampled-data adaptive PI control with uniformly bounded parameters

This paper studies sampled-data tracking control problems for first-order nonlinear time-invariant plants. A sampled-data adaptive PI controller is developed from exact discretization and full form dynamic linearization (FFDL) methods. To ensure the uniform boundedness of adaptive PI parameters with respect to sufficiently small sampling periods, novel lifted FFDL models and cost functions are introduced for designing controllers and adaptive rules. After establishing nonlinear closed-loop dynamics, new overall Lyapunov functions containing logarithmic operation are constructed for proving global stability and convergence. An extension to locally Lipschitz dynamics is given. Finally, two numerical examples and a practical application in longitudinal speed tracking for electrical cars are simulated to illustrate the efficiency and feasibility of the proposed results.