The maximum principle for stochastic control with partial information
U. G. Haussmann
Abstract
U. G. Haussmann
Abstract
In this paper a Pontryagin-type maximum principle is given for the following stochastic optimal control problem: the state of the system satisfies (i.e. is a weak solution of) an Ito equation with controlled drift and possibly degenerate diffusion coefficients; the controls available are functions of noise-corrupted observations of the state, and the cost to be minimized is the expected value of an integral cost plus a terminal cost. The proof of the Maximum Principle is given elsewhere; here we only state it carefully and then we apply it to the example of the Linear Regulator.
OpenAlex reports 1 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
In this paper a Pontryagin-type maximum principle is given for the following stochastic optimal control problem: the state of the system satisfies (i.e. is a weak solution of) an Ito equation with controlled drift and possibly degenerate diffusion coefficients; the controls available are functions of noise-corrupted observations of the state, and the cost to be minimized is the expected value of an integral cost plus a terminal cost. The proof of the Maximum Principle is given elsewhere; here we only state it carefully and then we apply it to the example of the Linear Regulator.
Key concepts: Maximum principle, Stochastic control, Optimal control, Degenerate energy levels, Pontryagin's minimum principle, Separation principle, Mathematics, Terminal (telecommunication)