Provably Safe and Robust Learning-Based Model Predictive Control
نویسندگان
چکیده
Controller design faces a trade-off between robustness and performance, and the reliability of linear controllers has caused many practitioners to focus on the former. However, there is renewed interest in improving system performance to deal with growing energy constraints. This paper describes a learning-based model predictive control (LBMPC) scheme that provides deterministic guarantees on robustness, while statistical identification tools are used to identify richer models of the system in order to improve performance; the benefits of this framework are that it handles state and input constraints, optimizes system performance with respect to a cost function, and can be designed to use a wide variety of parametric or nonparametric statistical tools. The main insight of LBMPC is that safety and performance can be decoupled under reasonable conditions in an optimization framework by maintaining two models of the system. The first is an approximate model with bounds on its uncertainty, and the second model is updated by statistical methods. LBMPC improves performance by choosing inputs that minimize a cost subject to the learned dynamics, and it ensures safety and robustness by checking whether these same inputs keep the approximate model stable when it is subject to uncertainty. Furthermore, we show that if the system is sufficiently excited, then the LBMPC control action probabilistically converges to that of an MPC computed using the true dynamics.
منابع مشابه
Improving the stability of the power system based on static synchronous series compensation equipped with robust model predictive control
Low-frequency oscillations (LFO) imperil the stability of the power system and reduce the Capacity of transmission lines. In the power systems, FACTS devices and Power System stabilizers are used to improve the stability. Static synchronous series compensators is one of the most important FACTS devices. This paper investigates the damping of LFO with static synchronous series compensator (SSSC)...
متن کاملDevelopment of RMPC Algorithm for Compensation of Uncertain Time-Delay and Disturbance in NCS
In this paper, a synthesis method based on robust model predictive control is developed for compensation of uncertain time-delays in networked control systems with bounded disturbance. The proposed method uses linear matrix inequalities and uncertainty polytope to model uncertain time-delays and system disturbances. The continuous system with time-delay is discretized using uncertainty po...
متن کاملLearning-based Model Predictive Control for Safe Exploration and Reinforcement Learning
Learning-based methods have been successful in solving complex control tasks without significant prior knowledge about the system. However, these methods typically do not provide any safety guarantees, which prevents their use in safety-critical, real-world applications. In this paper, we present a learning-based model predictive control scheme that provides provable high-probability safety gua...
متن کاملRobust Model Predictive Control for a Class of Discrete Nonlinear systems
This paper presents a robust model predictive control scheme for a class of discrete-time nonlinear systems subject to state and input constraints. Each subsystem is composed of a nominal LTI part and an additive uncertain non-linear time-varying function which satisfies a quadratic constraint. Using the dual-mode MPC stability theory, a sufficient condition is constructed for synthesizing the ...
متن کاملModelling and Compensation of uncertain time-delays in networked control systems with plant uncertainty using an Improved RMPC Method
Control systems with digital communication between sensors, controllers and actuators are called as Networked Control Systems (NCSs). In general, NCSs encounter with some problems such as packet dropouts and network induced delays. When plant uncertainty is added to the aforementioned problems, the design of the robust controller that is able to guarantee the stability, becomes more complex. In...
متن کاملRobust Learning Model Predictive Control for Uncertain Iterative Tasks: Learning From Experience
We present a Robust Learning Model Predictive Controller (RLMPC) for constrained uncertain systems performing iterative tasks. The proposed controller builds on earlier work of Learning Model Predictive Control (LMPC) for deterministic systems. The main idea behind RLMPC is to collect data from previous iterations and use it to estimate the current value function and build a robust safe set. We...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
- Automatica
دوره 49 شماره
صفحات -
تاریخ انتشار 2013