Iterative Linear Quadratic Regulator Design for Nonlinear Biological Movement Systems

Authors

Weiwei Li, Emanuel Todorov

Year

2004

Conference Paper

Algorithm

iLQR

Source

International Conference on Informatics in Control, Automation and Robotics (ICINCO)

Identifiers

DOI 10.5220/0001143902220229

Abstract¶

This paper presents an Iterative Linear Quadratic Regulator (ILQR) method for locally-optimal feedback control of nonlinear dynamical systems. The method is applied to a musculo-skeletal arm model with 10 state dimensions and 6 controls, and is used to compute energy-optimal reaching movements. Numerical comparisons with three existing methods demonstrate that the new method converges substantially faster and finds slightly better solutions.

Summary¶

Introduces iLQR, applying iterative linearization around the current nominal trajectory to reduce nonlinear trajectory optimization to a sequence of LQR problems, enabling efficient second-order trajectory optimization for nonlinear systems.