Title :
The δ - sensitivity and its application to stochastic optimal control of nonlinear diffusions
Author :
Theodorou, Evangelos A. ; Todorov, Emo
Author_Institution :
Dept. of Comput. Sci. & Eng., Univ. of Washington, Seattle, WA, USA
Abstract :
We provide optimal control laws by using tools from stochastic calculus of variations and the mathematical concept of δ-sensitivity. The analysis relies on logarithmic transformations of the value functions and the use of linearly solvable Partial Differential Equations(PDEs). We derive the corresponding optimal control as a function of the δ-sensitivity of the logarithmic transformation of the value function for the case of nonlinear diffusion processes affine in control and noise.
Keywords :
calculus; nonlinear control systems; optimal control; partial differential equations; stochastic systems; δ - sensitivity; PDE; linearly solvable partial differential equations; logarithmic transformations; mathematical concept; nonlinear diffusion processes; nonlinear diffusions; stochastic calculus; stochastic optimal control; Calculus; Cost function; Diffusion processes; Equations; Markov processes; Optimal control; Sensitivity;
Conference_Titel :
American Control Conference (ACC), 2013
Conference_Location :
Washington, DC
Print_ISBN :
978-1-4799-0177-7
DOI :
10.1109/ACC.2013.6580486
