DocumentCode :
1615914
Title :
Discrete mechanics modelling and real-time optimal control method based on Lie group variational integrator
Author :
Jie Li ; Xiangjiang Xiao ; Huayong Zhu ; Lincheng Shen
Author_Institution :
Coll. of Mechatron. Eng. & Autom., Nat. Univ. of Defense Technol., Changsha, China
fYear :
2013
Firstpage :
930
Lastpage :
937
Abstract :
Under discrete mechanics, a real-time optimal control method is developed based on Lie group variational integrator. First, the Hamilton-Pontryagin principle is promoted as a variational principle called the d´ Alembert-Pontryagin principle by introducing virtual displacement and virtual work, that would be applied to the mechanical system with nonconservative force and control. The left-trivialized d´ Alembert-Pontryagin equation of a controlled mechanical system is derived in continuous time and discrete time, respectively. Second, the first-order necessary optimality condition of a discrete-time optimal control problem is derived by using discrete variation method on Lie group, and thus a two-point boundary value problem is created. Then, referring to C/FD-GMRES method, an algorithm is developed based on receding horizon control for solving the two-point boundary value problem. Finally, the validity of the presented algorithm is verified by simulation example of optimal control problem on SE(2).
Keywords :
Lie groups; boundary-value problems; continuous time systems; discrete time systems; integration; nonlinear control systems; optimal control; variational techniques; C-FD-GMRES method; Hamilton-Pontryagin principle; Lie group variational integrator; continuation-forward difference-generalized minimum residual method; continuous time system; d´ Alembert-Pontryagin principle; discrete mechanics modelling; discrete time system; discrete variation method; discrete-time optimal control problem; first-order necessary optimality condition; mechanical system; nonconservative force; realtime optimal control method; receding horizon control; two-point boundary value problem; virtual displacement; virtual work; Equations; Force; Hafnium; Mathematical model; Mechanical systems; Optimal control; Lie group; discrete mechanics; optimal control; receding horizon control; variational integrator;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Chinese Automation Congress (CAC), 2013
Conference_Location :
Changsha
Print_ISBN :
978-1-4799-0332-0
Type :
conf
DOI :
10.1109/CAC.2013.6775865
Filename :
6775865
Link To Document :
بازگشت