Iterative linearization methods for approximately optimal control and estimation of non-linear stochastic system

Abstract

This paper presents an iterative Linear-Quadratic-Gaussian method for locally-optimal control and estimation of non-linear stochastic systems. The new method constructs an affine feedback control law, obtained by minimizing a novel quadratic approximation to the optimal cost-to-go function, and a non-adaptive estimator optimized with respect to the current control law. The control law and filter are iteratively improved until convergence. The performance of the algorithm is illustrated on a complex biomechanical control problem involving a stochastic model of the human arm.