Flexible-step Model Predictive Control based on Generalized Lyapunov Functions

We present a novel nonlinear model predictive control (MPC) scheme with relaxed stability criteria, based on the idea of generalized discrete-time control Lyapunov functions. These functions need to satisfy an average descent over a finite window of time, rather than a descent at every time step. One feature of this scheme is that it allows for implementing a flexible number of control inputs in each iteration, in a computationally attractive manner, while guaranteeing recursive feasibility and stability. The benefits of our flexible-step implementation are also demonstrated in an application to nonholonomic systems, where the one-step standard implementation may suffer from lack of asymptotic convergence.