Nonlinear receding horizon gradient method- real-time optimization for robot trajectory generation

The proposed method reduces a hardness to treat additional formula manipulation in past receding horizon control. Historically, gradient method had a defect to take long time to converge into optimal solution. This defect can be turned into an advantage by taking different procedure to use gradient. This paper describes how to use gradient at receding horizon control. This method is also effective for a case which the state equation is complicated or large scaled model because the formulation is simple.

[1]  Toshiyuki Ohtsuka,et al.  Real-time optimization algorithm for nonlinear receding-horizon control , 1997, Autom..

[2]  Toshiyuki Ohtsuka,et al.  Continuation/GMRES method for fast algorithm of nonlinear receding horizon control , 2000, Proceedings of the 39th IEEE Conference on Decision and Control (Cat. No.00CH37187).

[3]  J. Doyle,et al.  On receding horizon extensions and control Lyapunov functions , 1998, Proceedings of the 1998 American Control Conference. ACC (IEEE Cat. No.98CH36207).

[4]  Hiroki Takeuchi.  Real Time Optimization for Robot Control using Receding Horizon Control with Equal Constraint , 2003, J. Field Robotics.