Stochastic Model-Predictive Control of Uncertain System
AssigneeMitsubishi Electric Research Laboratories, Inc.
InventorsKarl Berntorp, Rien Quirynen, Sean Vaskov
Life of this patent
Estimating the term…
Abstract
A stochastic model predictive controller (SMPC) estimates a current state of the system and a probability distribution of uncertainty of a parameter of dynamics of the system based on measurements of outputs of the system, and updates a control model of the system including a function of dynamics of the system modeling the uncertainty of the parameter with first and second order moments of the estimated probability distribution of uncertainty of the parameter. The SMPC determines a control input to control the system by optimizing the updated control model of the system at the current state over a prediction horizon and controls the system based on the control input to change the state of the system.
Claims19
A stochastic model predictive controller (SMPC) for controlling a system, comprising:
- at least one processor;
- and memory having instructions stored thereon that, when executed by the at least one processor, cause the SMPC to:
- determine, at each control step based on measurements of outputs of the system indicative of a transition of a state of the system from a previous state to a current state, an estimation of the current state of the system, and an estimation of a probability distribution of uncertainty of a parameter of dynamics of the system;
- update a control model of the system including a function of dynamics of the system modeling the uncertainty of the parameter with first and second order moments of the estimated probability distribution of uncertainty of the parameter;
- determine a control input to control the system by optimizing the updated control model of the system at the current state over a prediction horizon;
- and
- control the system based on the control input to change the state of the system.
The SMPC of claim 14, wherein the probability distribution of uncertainty of the parameter is modeled as a Gaussian distribution with unknown mean and covariance estimated by the Kalman filter during the control of the system as the first and the second order moments.
A stochastic model predictive control (SMPC) method for controlling a system, wherein the method uses a processor coupled with stored instructions implementing the method, wherein the instructions, when executed by the processor carry out steps of the method, comprising:
- determining, at each control step based on measurements of outputs of the system indicative of a transition of a state of the system from a previous state to a current state, an estimation of the current state of the system, and an estimation of a probability distribution of uncertainty of a parameter of dynamics of the system;
- updating a control model of the system including a function of dynamics of the system modeling the uncertainty of the parameter with first and second order moments of the estimated probability distribution of uncertainty of the parameter;
- determining a control input to control the system by optimizing the updated control model of the system at the current state over a prediction horizon;
- and
- controlling the system based on the control input to change the state of the system.
Dependencies are inferred from the claim text.