Iterative Learning Control based on Stochastic Approximation ⋆
نویسندگان
چکیده
In this paper stochastic approximation theory is used to produce Iterative Learning Control (ILC) algorithms which are less sensitive to stochastic disturbances, a typical problem for the learning process of standard ILC algorithms. Two algorithms are developed, one to obtain zero mean controlled error and one to minimise the mean 2-norm of the controlled error. The former requires a certain knowledge of the system but in the presence of noise can give reasonably rapid convergence. The latter can either use a model or be model free by employing a second experiment.
منابع مشابه
Stochastic Point-to-point Iterative Learning Control Based on Stochastic Approximation
An iterative learning control algorithm with iteration decreasing gain is proposed for stochastic point-to-point tracking systems. The almost sure convergence and asymptotic properties of the proposed recursive algorithm are strictly proved. The selection of learning gain matrix is given. An illustrative example shows the effectiveness and asymptotic trajectory properties of the proposed approach.
متن کاملApproximate Inference and Stochastic Optimal Control
We propose a novel reformulation of the stochastic optimal control problem as an approximate inference problem, demonstrating, that such a interpretation leads to new practical methods for the original problem. In particular we characterise a novel class of iterative solutions to the stochastic optimal control problem based on a natural relaxation of the exact dual formulation. These theoretica...
متن کاملLearning Algorithms for Risk-Sensitive Control
This is a survey of some reinforcement learning algorithms for risk-sensitive control on infinite horizon. Basics of the risk-sensitive control problem are recalled, notably the corresponding dynamic programming equation and the value and policy iteration methods for its solution. Basics of stochastic approximation algorithms are also sketched, in particular the ‘o.d.e.’ approach for its stabil...
متن کاملStochastic Shortest Path Games and Q-Learning
We consider a class of two-player zero-sum stochastic games with finite state and compact control spaces, which we call stochastic shortest path (SSP) games. They are total cost stochastic dynamic games that have a cost-free termination state. Based on their close connection to singleplayer SSP problems, we introduce model conditions that characterize a general subclass of these games that have...
متن کاملOn Probabilistic Inference Approaches to Stochastic Optimal Control
While stochastic optimal control, together with associate formulations like Reinforcement Learning, provides a formal approach to, amongst other, motor control, it remains computationally challenging for most practical problems. This thesis is concerned with the study of relations between stochastic optimal control and probabilistic inference. Such dualities – exemplified by the classical Kalma...
متن کامل