Advantages of Di erential Dynamic Programming OverNewton ' s Method for Discrete - Time Optimal Control
نویسنده
چکیده
Diierential Dynamic Programming (DDP) and stagewise Newton's method are both quadratically convergent algorithms for solving discrete time optimal control problem. Although these two algorithms share many theoretical similarities, they demonstrate signii-cantly diierent numerical performance. In this paper, we will compare and analyze these two algorithms in detail, and derive another quadratically convergent algorithm which is a combination of the DDP algorithm and Newton's method. This new second-order algorithm plays a key role in the explanation of the numerical diierences between the DDP algorithm and Newton's method. The detailed algorithmic and structural diierences for these three algorithms and their impact on numerical performance will be discussed and explored. Two test problems with various dimensions solved by these three algorithms will be presented. One nonlinear test problem demonstrates that the DDP algorithm can be as much as 28 times faster than the stagewise Newton's method. The numerical comparison indicates that the DDP algorithm is numerically superior to the stagewise Newton's method.
منابع مشابه
Extracting Dynamics Matrix of Alignment Process for a Gimbaled Inertial Navigation System Using Heuristic Dynamic Programming Method
In this paper, with the aim of estimating internal dynamics matrix of a gimbaled Inertial Navigation system (as a discrete Linear system), the discretetime Hamilton-Jacobi-Bellman (HJB) equation for optimal control has been extracted. Heuristic Dynamic Programming algorithm (HDP) for solving equation has been presented and then a neural network approximation for cost function and control input ...
متن کاملDiscrete Time Control Method for SVM Direct Active Power and Stator Flux Control of PMSG-Based Wind Turbine
This paper proposes a new method for direct control of active power and stator flux of permanent magnet synchronous generator (PMSG) used in the wind power generation system. Active power and stator flux are controlled by the proposed discrete time algorithm. Despite the commonly used vector control methods, there is no need for inner current control loops. To decrease the errors between refere...
متن کاملA numerical approach for optimal control model of the convex semi-infinite programming
In this paper, convex semi-infinite programming is converted to an optimal control model of neural networks and the optimal control model is solved by iterative dynamic programming method. In final, numerical examples are provided for illustration of the purposed method.
متن کاملAdaptive-critic based optimal neuro control synthesis for distributed parameter systems
A neural network based optimal control synthesis approach is presented for systems modeled by partial di!erential equations. The problem is formulated via discrete dynamic programming and the necessary conditions of optimality are derived. For synthesis of the controller, we propose two sets of neural networks: the set of action networks captures the mapping between the state and control, while...
متن کاملNumerical Solution of fuzzy differential equations of nth-order by Adams-Bashforth method
So far, many methods have been presented to solve the rst-order di erential equations. But, not many studies have been conducted for numerical solution of high-order fuzzy di erential equations. In this research, First, the equation by reducing time, we transform the rst-order equation. Then we have applied Adams-Bashforth multi-step methods for the initial approximation of one order di erentia...
متن کامل