CONTRACT-BASED MULTI-RATE CONTROL DESIGN

Abstract

This thesis investigates a contract-based control design framework that aims to solve the constraint partition problem raised in multi-rate control. There are three components in such a framework: the contract coordinator, the high-level motion planner, and the low-level tracking controller. The primary idea of the contract coordinator is to resolve the online constraint partition problem, thereby enhancing coordination between levels and enabling a more adaptable multi-rate control framework. This flexibility in the control framework translates to more choices in selecting the initial state. The contract coordinator is designed based on solving a linear matrix inequality (LMI) problem in real-time, which takes the online information from both levels and outputs the new constraint partition online. Given the input and state constraints, a robust model predictive control with varying constraints is proposed for high-level motion planning. The low level implements the robust control barrier function-based quadratic programming to ensure safety given the constraints from the LMI coordinator. The theoretical guarantees of each level, including feasibility, safety, and stability, are provided. One numerical simulation example is given to show the effectiveness of the proposed method.