Efficient particle continuation model predictive control

TL;DR

This paper introduces a particle continuation MPC method that efficiently handles systems with discrete changes in dynamics or constraints by computing ensembles of controls online, improving adaptability and optimality in real-time control scenarios.

Contribution

It proposes a novel particle continuation MPC algorithm for online control of systems with changing dynamics, demonstrating its effectiveness through numerical experiments.

Findings

Effective handling of systems with discrete dynamic changes.

Real-time computation of control ensembles for multiple scenarios.

Improved adaptability and optimality in control tasks.

Abstract

Continuation model predictive control (MPC), introduced by T. Ohtsuka in 2004, uses Krylov-Newton approaches to solve MPC optimization and is suitable for nonlinear and minimum time problems. We suggest particle continuation MPC in the case, where the system dynamics or constraints can discretely change on-line. We propose an algorithm for on-line controller implementation of continuation MPC for ensembles of predictions corresponding to various anticipated changes and demonstrate its numerical effectiveness for a test minimum time problem arriving to a destination. Simultaneous on-line particle computation of ensembles of controls, for several dynamically changing system dynamics, allows choosing the optimal destination on-line and adapt it as needed.